electrochemical immunosensors and peptide self …tesi.cab.unipd.it/26611/1/solda_tesi.pdf ·...

UNIVERSITÀ DEGLI STUDI DI PADOVA

Facoltà di Scienze MM. FF. NN.

Dipartimento di Scienze Chimiche

Tesi di Laurea Magistrale in Chimica

Electrochemical Immunosensors and Peptide Self-Assembled Monolayers

for Cancer Biomarker Protein Detection

Relatore: Prof. Flavio Maran

Correlatore: Prof. James F. Rusling (University of Connecticut, USA)

Controrelatore: Prof. Fabrizio Mancin

Laureando: Alice Soldà

Anno Accademico 2009-2010

3

“Non basta guardare,

occorre guardare con occhi che vogliono vedere, e che credono in quello che vedono.”

(Galileo Galilei)

5

Acknowledgments

I would like to thank Prof. Flavio Maran for believing in me since the first

time. He gave me the opportunity to complete my Thesis work in the University

of Connecticut (UCONN, Storrs, CT). Most importantly, I would like to thanks

him for his moral and financial support, and for the thousands of good advice that

helped me to grow professionally and temperamentally in these months.

I would like to thank Prof. James Rusling, (Department of Chemistry of

the University of the Connecticut – UCONN, Storrs, CT) for his great hospitality,

the financial support and the opportunity that he gave me to work in his research

group: in those months, I learnt many things about the immunosensors’ world.

I would like also to thanks Dr. Dharamainder Chaudary (UCONN Health

Center, Farmington, CT) to provide me Nanog biomarker samples and the relative

antibodies and Prof. Edward Samulski (Department of Chemistry, University of

North Carolina, Chapel Hill, NC) for clearing to all my doubts about IRRAS

investigation, even though he did not know me directly.

I am grateful to all members of Prof. Maran’s group: from Dr. Sabrina

Antonello, who constantly helped and supported me even in the most critical

moments with good advices, to Martina for her friendship and for making more

pleasant the hard lab’s days with her songs and Ivan, for providing me all

peptides, even in relatively short times. A special thank is addressed to

Pierangelo, which initially convinced me to take this road and after passed me his

knowledge about SAMs’ procedure ... (lo ringrazio un po’ meno per le costanti

battutine a doppio senso con cui mi perseguitava ogni giorno).

I am grateful also to all members of Prof. Rusling’s group that followed

and helped me during the period that I spent there. In particular Bhaskara, for

teaching me everything I know now about the biosensors (although sometimes the

experiments followed more a mystical way, that a scientific one with all those

“Abrakadabra”), and Ruchika for her support and advices. I would say also

thanks to Ahbay, Danuka, Vigneshwaran, Alex, Linlin, Collen, Sadagopan,

Niamesh, Shenmin, and the others.

7

Index

Chapter 1. Introduction .............................................................................. 9

1.1 Biosensors ........................................................................................ 9

1.1.1 Electrochemical Immunosensors ............................................. 11

1.1.2 Microfluidic Electrochemical Device for High Sensitivity

Biosensing ............................................................................... 13

1.1.3 Nanog Protein as Biomarker of Cancer .................................. 14

1.2 Self-Assembled Monolayers (SAMs) as a Platform for

Biosensors ..................................................................................... 16

1.2.1 Electron Transfer through SAMs ............................................ 17

1.2.2 Self-Assembled Monolayer .................................................... 19

1.2.3 Types of Substrates and Characterization of SAMs ................ 21

1.2.4 Nature of the Metal-SAM Interface ......................................... 23

1.2.5 Organization of the Organic Layer .......................................... 25

1.3 Gold Electrochemical Biosensors ............................................ 27

1.4 α-Aminoisobutyric Acid 310-Helices as Adsorbate for

SAMs............................................................................................... 28

1.4.1 Characteristic of the Investigated Aib-Homopeptides ............. 33

1.5 Purposes of the Thesis ................................................................ 36

Chapter 2. Experimental .......................................................................... 38

2.1 Chemicals ....................................................................................... 38

2.2 Preparation of HRP Single Electrodes .................................... 38

2.3 Fabrication of the Microfluidic Device and Preparation of

Eight-Electrode Array ................................................................. 42

8

2.4 Gold Preparation and SAM Formation .................................. 44

2.5 Instruments .................................................................................... 46

2.5.1 Electrochemistry ...................................................................... 46

2.5.2 IRRAS ..................................................................................... 46

Chapter 3. Results and Discussion .................................................... 47

3.1 Electrochemical Immunosensors ............................................. 47

3.1.1 Characterization of AuNP Platform ....................................... 47

3.1.2 Electrochemistry Immunosensors ........................................... 50

3.1.3 Analysis of the Immunosensors’ Performance........................ 53

3.2 IRRAS Characterization of Peptide SAM ............................. 69

3.2.1 The Metal-Surface Selection Rule .......................................... 70

3.2.2 Characterization of Aib-Homopetides .................................... 74

3.2.3 Characterization of Aib-Homopeptide SAMs ......................... 79

3.2.4 Determination of the Orientation of Adsorbed Molecules ...... 93

3.2.5 Determination of the Thickness of the Monolayer ................ 100

Chapter 4. Conclusions ......................................................................... 103

Chapter 5. References ............................................................................ 107

11

1. Introduction

1.1 Biosensors

The IUPAC definition of a chemical sensor is a: “a device that transforms

chemical information, ranging from the concentration of a specific sample

component to total composition analysis, into an analytically useful signal. The

chemical information may originate from a chemical reaction of the analyte or

from a physical property of the system investigated. A chemical sensor is an

essential component of an analyzer. In addition to the sensor, the analyzer may

contain devices that perform the following processing. Chemical sensors contain

two basic functional units: a receptor part and a transducer part. In the receptor

part of a sensor, the chemical information is transformed into a form of energy

that may be measured by the transducer. The transducer part is a device capable of

transforming the energy carrying the chemical information about the sample into a

useful analytical signal”.1

“A biosensor is a particular kind of chemical sensor that uses specific

biochemical reactions mediated by isolated enzymes, immunosystems, tissues,

organelles or whole cells to detect chemical compounds usually by electrical,

thermal or optical signal”.1 Nowadays the interest about biosensors is rapidly

increasing especially in the fields of health care, food and environmental quality.

A representative example is provided by the glucose biosensor, which is

used by millions diabetic people all over the world. Now people can make by

themselves a fast control of their glucose level everywhere and, depending on the

output of the sensor, they can inject the exact amount of insulin to bring the

glucose to an optimal level.2

For these reasons, economic investments aimed to supporting the

biosensor research are increasing. The technology behind biosensors is wide and

multidisciplinary, ranging from biology, chemistry, engineering and electronics,

all these disciplines being necessary to create a well performing biosensor.3,4

12

A biosensor is made by two essential components integrated: a bioreceptor

and a transducer. The first one is a biological molecule that can specifically

interact and recognize the molecules present in the sample analyte, for example

proteins, DNA, enzymes, but also a whole cell. The second component is an

electrochemical or optical device that converts the recognition event into a

measurable signal (Figure 1).

Figure 1. General biosensor scheme.

The outcome is both qualitative and quantitative because while the

specificity of the signal is guaranteed by the bioreceptor, the intensity is most

often related to the concentration of the analyte. From these considerations, we

infer the most important features that a good biosensor should possess:

I. The bioreceptor must have a very high specificity toward the analyte, a

good stability under different experimental conditions, and a good

reproducibility;

II. The transducer must have good properties as either an electron transfer

mediator or an optical device, and it should be stable and easily modified

or functionalized;

III. To allow for a fast and easy utilization, the pre-treatment phase should be

minimal. It also is important to keep at minimum the number of

parameters that could influence the performance of the measurement;

IV. The output should be accurate, easily understandable, reproducible, free of

background noise, and in the range of interest;

V. It should be cheap and user friendly.

13

Based on the position of the transducer, there are three main types of

biosensors; the products of the reactions diffuse to the transducer causing an

electrical signal; the products of the reactions are identified by a mediator that

interacts with the transducer to generate the signal; the reaction itself is

recognized by the transducer. The transducer can then be classified according to

the mechanism of analyte recognition. Electrochemical, optical, and thermal

biosensors can be identified based on the type of signal used for gathering

information about the analyte.5

Electrochemical biosensors can be classified according to the specific

electrochemical method employed. A potentiometric biosensor is based on the

measurement of variations of the electrode potential with respect to a reference

electrode, or the potential difference between two references electrodes separated

by a permeoselective membrane.6 An amperometric biosensor measures the

current generated by an applied potential between two electrodes.

Optical biosensors are based on surface plasmon resonance, and they have

recently received considerable attention owing to their high accuracy and

reliability. They measure the different ways by which materials or molecule may

interact with the light (absorbance of the light, the reflected light, the light output

by a luminescent reaction, etc.). One of the most popular materials used in this

field is gold nanoparticles of suitable size.7

Thermal (calorimetric) biosensors measure the difference of temperature

before and after the reaction. In fact, the majority of the enzymes catalyze

exothermic reactions and thus the heat generated can be measured.8

1.1.1 Electrochemical Immunosensors

Sensitive quantitative detection of protein biomarkers is critical to many

areas of biomedical research and diagnostics, systems biology, and proteomics.

Biomarker levels in serum can detect and monitor diseases such as cancer. In this

context, sensors need to be simple operationally, capable of rapid multiplexed

detection, inexpensive, and must display sufficiently good sensitivity and

detection limits to address the levels of the biomarkers in both normal and cancer

patient serum.9

14

Conventional ways of measuring proteins include enzyme-linked

immunosorbent assays (ELISA), radioimmunoassay (RIA), electrophoretic

immunoassay, and mass spectroscopy-based proteomics. These techniques often

involve sophisticated instrumentation, significant sample volumes, limited

sensitivity and clinically unrealistic expense and time. ELISA-like approaches

have been successfully adapted to immunoarrays systems. These ultrasensitive

multilayer arrays, relying on optical or electrical detection, have considerable

promise for achieving point-of-care measurement.9

Immunosensors are biosensors based on the antigen-antibody interaction,

which is responsible for eventually generating the actual signal. This type of

biosensors has, in principle, high specificity and low limit of detection thanks to

the extreme affinity that antibodies have for their antigen. Whereas antibodies are

proteins produced by the immune system, antigens can be a variety of different

molecules, from protein to DNA, lipids, etc. 10 In the following, we provide a brief

description of the structure of antibodies to appreciate better how they work and

can be used in biosensors.

Antibodies are heavy globular plasma proteins (150 kDa). They are also

called glycoproteins because they are composed by sugar chains and amino acid

residues (Figure 2). The basic functional unit of each antibody is an

immunoglobulin monomer, but it can be also dimeric, tetrameric or pentameric. 11

Figure 2. Antibody’s structure cell.

15

The monomer has a “Y” form, which is made of 4 chains: two identical

light and two identical heavy chains. These chains are connected by disulfide

bonds. Both chains are also divided in two regions: one constant and one variable.

The variable regions of both the heavy and of the light chains interact to form the

antigen binding site, and thus each monomer has two sites with the same

specificity to recognize and link the antigen. The constant region of the heavy

chains determines the function of the antibody.11

Immunosensors are most often optical or electrical and therefore antigens

and antibodies themselves are not sufficient to generate a sizeable detection

signal. For this reason, to permit the transfer of the signal it is necessary to

conjugate a label molecule or a material to the first antibody; gold nanoparticles

are often used for their particular properties. The chemical groups of the antibody

used for conjugations are mainly amine (-NH2), thiol (-SH), or hydroxyl (-OH)

groups.12-13

Of particular relevance to the work carried out during this Laurea

Magistrale thesis, which based on electrochemical biosensors, is the presence of a

redox enzyme. To increase the electric signal, a secondary antibody (Ab2) is often

added. Its function is to bind both to the analyte protein/s and to the redox

enzyme, such as horseradish peroxidase (HRP). Rusling and co-workers found

that, there are 14-16 HRP units per secondary antibody.9 On the other hand, the

same authors found that sensitivity is greatly amplified by using magnetic beads

bioconjugated with HRP labels, the number of active HRP per nanoparticle being

estimated to be 7500.9

1.1.2 Microfluidic Electrochemical Device for High

Sensitivity Biosensing

Another aspect relevant to my Thesis concerns microfluidic devices. The

latter have the ability to analyze very small quantities of sample, to limit the

reagent use, and to carry out analyses at high resolution and sensitivity, low cost,

and in short time. Microfluidic devices have applications in biology, chemistry

and medicine, including measurement of diffusion coefficient, fluid viscosity, and

16

binding constant, as well as DNA analysis, cell separation, cell pattering, capillary

electrophoresis and immunoassays. Implementing biosensors in microfluidic

format provides a potentially more efficient approach to control and automate

sample introduction and steps such as washing and reagent addition. By coupling

with microfluidics, immunoassay procedures could potentially be made fast

enough for point-of-care without sacrificing sensitivity.14

Microfluidics devices are commonly fabricated in glass, silicon, or

polymers, with polymers finding considerable recent attention.

Poly(dimethylsiloxane) (PDMS) is used extensively to fabricate these

microfluidic devices using photolithography or more simply by polymer

deposition onto molds. The work carried out in the Rusling group at the

University of Connecticut, and relevant to this Thesis research, relies on an

electrochemical sensor constructed by integrating injection-molded electrodes into

a polystyrene micro-flow channel.14 A simple microfluidic device was used for

electrochemical biosensing, fabricated by mold deposition and validated by

sensitive detection of hydrogen peroxide. The device features a single

microfluidic channel made from PDMS coupled to a fixed volume injector, and

incorporates a biocatalytic sensing electrode, a reference electrode and a counter-

electrode. PDMS was chosen because it is easily moldable, soft and readily

integrated with outside components, and effective protocols exist to inhibit

biomolecular contamination. Whereas the detection limits of biomarkers in a

microfluidic biosensor are usually better than those obtained with a simple

rotating-disk electrode in a conventional electrochemical cell, the sensitivity is

comparable between the two systems. The improvement of detection limit may be

related to a better control of mass transport in the microfluidic system compared

to the rotating-disc electrode system, leading to better signal-to-noise.14

1.1.3 Nanog Protein as Biomarker of Cancer

A few words about the specific biomarker protein used in my Thesis work

are necessary. The stem-cell-abundant protein Nanog is highly expressed in

undifferentiated embryonic stem (ES) cells and regulates stem-cell differentiation.

Nanog is a unique homeobox transcription factor and has a homeodomain with

17

homology to members of the natural killer gene family; indeed it has a similar

critical role in regulating the cell fate of the pluripotent ICM (inner cell mass)

during embryonic development, maintaining the pluripotency epiblast and

preventing differentiation.15

This protein may play a role in carcinogenesis of embryonic cancer,

gliomas, liver cancer, gastric cancer, and other cancers. The role of Nanog in the

transformation of cervical epithelial cells carcinoma, and the occurrence and

development of cervical carcinoma have not been investigated in detail. The

expression of Nanog in cervical epithelia lesions of varying severity and in

cervical carcinoma by immunohistochemical analysis have been addressed.15 The

specimens were obtained from 253 patients: of these 49 had a normal cervical

epithelia, 31 had mild dysplasia (CIN I), 77 had moderate-severe-dysplasia (CIN

II-III) and 78 had squamous cervical carcinomas (SCC). They found that the

expression levels of Nanog were higher in sample from SCC patients than in

samples from patients with normal cervical epithelia and CIN; they were also

higher in samples from patients with CIN than from those with normal cervical

epithelia. Nanog expression levels showed also significant differences according

to different tumor sizes (Figure 4).15

Nanog

Total N = 235

0 (-) N = 26

1 (+) N = 85

2 (++) N = 76

3 (+++) N = 48

P

Normal 49 17 22 8 2 0.000 CIN I 31 2 18 9 2 0.009 CIN II-III SCC

77 78

4 3

25 20

35 24

13 31

0.015 0.017

Figure 3. Expression of Nanog in cervical epithelial lesions of varying severity:

- 0(-) = < 5% positive cells;

- 1(+) = 5-25% positive cells;

- 2(++) = 26-75% positive cells;

- 3(+++) = more than 76% positive cells.

18

Figure 4. Nanog expression and localization in (A) normal cervical epithelial

cells; (B) CIN I cells; (C) CIN II-III cells; (D) SCC cells. The distinct brown color

indicative of Nanog was detected in the cytoplasm of the positive cells.

1.2 Self-Assembled Monolayers (SAMs) as

Platforms for Biosensors

The building of biochemical sensors and, particularly, electrochemical

biosensors based on the deposition of functionalized layers on solid substrates is

gaining increasing importance.16 Making surfaces that can be used for biosensing

involves surface modifications aimed at specifically changing the way they

interact with the environment, usually a solution. Biosensors are devised to bind

the analyte and transduce the binding event to a sizeable output signal, whether

optical or electrochemical, that can be used for quantification. The nature of the

surface is of paramount importance to determine the performance of the device or

sensor. In electrochemical biosensors, where detection is associated with the onset

of a reduction or an oxidation current, the second important ingredient for

devising a well-performing system is the efficiency of electron transfer (ET).

Understanding and controlling electron conduction through the monolayer (or

multilayer) interposed between the underlying conducting substrate and solution

species diffusing in proximity of the outer monolayer interphase is indeed

essential for transducing the biomolecule recognition into a significantly large

19

electrochemical current. In recent years, several groups have been studying ET

reactions through metal electrode – organic monolayer systems, particularly those

based on self-assembled monolayers (SAMs) obtained by spontaneous adsorption

of thiolated molecules or disulfides onto gold substrates.16b,17 Main factors

affecting the ET rate through SAMs are the monolayer thickness, the structure and

orbitals of the adsorbed thiols, the presence of ionic or pH-sensitive terminal

groups on the solution side, and further factors related to the solution.

1.2.1 Electron Transfer through SAMs

Understanding the mechanisms of ET reactions is a fundamental challenge

in a variety of areas of chemistry and biochemistry.18-21 Considerable

achievements in this direction have been gathered by studying long-range electron

or hole transfers in proteins20a and DNA.21a Long-range ET reactions are

intrinsically nonadiabatic and may proceed by different mechanisms. One

important path for long-range ET is provided by the superexchange mechanism,

where the electron tunneling is mediated by the bridge separating the donor and

the acceptor but without transient occupation of the bridge electronic states.19a,22

The ET rate constant depends on the electronic coupling between the reactant and

product states at the transition state and is proportional to exp(−βdDA), where dDA

is the donor-acceptor distance and β is the exponential factor describing the falloff

rate through the specific bridge. Alternatively, the bridge may provide localized or

partially delocalized electronic states where electrons may hop by an incoherent

mechanism.21a,23 Because electron injection into the first bridge unit is the slow

step, the ET rate is mildly dependent on the increase of bridge units. Sequential

electron hopping may become more efficient than superexchange when the bridge

is made sufficiently long. Such a competitive-reaction scheme has received a

general consensus for charge transfer across DNA strands,21a,23a but it has not been

clearly assessed whether this scheme can be extended to proteins and thus

peptides,24 unless suitable amino-acid side-chain groups are present along the

peptide chain.25

Long-range ETs have been extensively studied, using freely diffusing

donor – molecular bridge – acceptor systems (with either outer-sphere19a or

20

dissociative-type acceptors26), electrode – molecular bridge – electrode

junctions,27 and electrode – SAM – redox moiety, where the latter (e.g., a

ferrocenyl group or a redox protein) is covalently or electrostatically bounded to

the solution side of the SAM’s adsorbate/s.17a,17b,17e,28-32 The distance dependence

of the ET rate and thus the observed β factor depend on the actual molecule

forming the monolayer and thus the nature of the bonds of the bridge separating

donor and acceptor. With alkanethiol ferrocene-terminated SAMs, where only

saturated C-C bonds are present, the dependence is exponential, in agreement with

the superexchange mechanism.28 On the other hand, different outcomes have been

reported for peptide SAMs decorated by ferrocene moieties. Whereas some

peptide SAMs display a simple exponential dependence, though with β values

remarkably smaller than for hydrocarbon chains, there are reports of the

observation of very mild distance dependences of the ET rate (β < 0.1 Å-1) that

were interpreted as due to a hopping ET mechanism or to the dynamics of the of

α- or β-amino acid peptide chain.29,30a,33

The number of ET studies based on the determination of ET rate constants

between SAM-modified electrodes and freely diffusing solution species is more

limited31a,34-39 and, in fact, no study has been yet reported for peptide SAMs.

Indeed, the use of soluble redox probes is often employed to test how tight and

blocking is a given SAM. Generally, the use of solution redox probes is affected

by problems in assessing and controlling the quality and packing of the SAMs

(see below), which implies a possible penetration of the redox species through the

SAM defects and thus an apparent increase of the ET rate. The effect of having

charges on the SAM periphery or different pH values and electrolytes is solution

has been addressed. Electrostatic repulsion between charged head groups in the

outer monolayer periphery and charged probes influences the ET rate quite

significantly. For example, alkanethiols with positive terminal functionalized

groups and self-assembled on gold can block ET between the substrate and a

positively charged electroactive species dissolved in the electrolyte solution.40

Most studies, however, have been carried out with uncharged monolayers and

constant solvent/electrolyte conditions, and thus imply no coulombic corrections

in the definition of the ET rate constant.

21

Beside intrinsic factors directly related to the chemical nature of the

molecules forming the SAM, there are important issues concerning the dynamics

of the SAM41 and the quality of the latter. Indeed, although making SAMs

involves relatively simple chemistry, the resulting modified surfaces are affected

by heterogeneity in coverage and this may affect signal transduction in

electrochemical sensors. For example, by using a fluorescence-microscopy

electrochemical method, Bizzotto and co-workers studied the potential-dependent

desorption of alkanethiolate and DNA SAMs on gold surfaces, and concluded that

heterogeneity in surface coverage may be stronger than expected.42 Indeed, the

main reason why long-range ET through SAMs has been studied mostly by using

redox groups linked to the end of the monolayer adsorbate is that this strategy

helps minimizing the problem of defects, as the presence of the latter should not

affect (at least for sufficiently well-packed SAMs) the main electron tunneling

pathway, i.e., through bonds. On the other hand, transduction in sensors involves

detection of redox species dissolved in solution and thus a preliminary screening

of the ET mediating property of a given SAM needs to be based on the

observation of the electrochemical behavior of a freely diffusing redox probe. ET

occurs by electron tunneling through the SAM but also through pinholes and other

defects. The observed ET rate constant is thus generally the result of a

combination of contributions and discerning between them is a difficult

task.34,43,44

1.2.2 Self-Assembled Monolayer

SAM form from spontaneous adsorption of a molecular layer onto a

substrate. Bare surfaces of metals and metal oxides tend to accidentally adsorb

organic materials readily because these adsorbates lower the free energy of the

interface between the metal or metal oxide and the environment. These adsorbates

also alter interfacial properties and can have a significant influence on the stability

of nanostructures of metals and metal oxides. The organic material can act as a

physical or electrostatic barrier against aggregation, decreasing the reactivity of

the surface atoms, or act as an electrically insulating film. Surfaces coated with

these materials, however, are not well defined as do not present specific chemical

22

functionalities and do not have reproducible physical properties (e.g.,

conductivity, wettability, or corrosion resistance). Nevertheless, SAMs provide a

convenient, flexible, and simple system with which to tailor the interfacial

properties of metals, metal oxides, and semiconductors.45-55

SAMs form by the adsorption of molecular constituents from solution or

gas phase onto the surface of solids. The adsorbates organize spontaneous into

crystalline or semicrystalline structures.56 The molecules (or ligands) that form

SAMs have a chemical functionality, or “head group”, with a specific affinity for

a substrate. There are a number of head groups that bind to specific metals, metal

oxides, and semiconductors, but the most extensively studied class of SAMs is

derived from the adsorption of alkanthiols on gold. The high affinity of thiols for

the surfaces of noble and coinage metals makes it possible to generate well-

defined organic surfaces with useful and highly alterable chemical functionalities

displayed at the exposed interface. SAMs are therefore nanostructures with a

number of useful properties. The composition of the molecular components of the

SAM determines the atomic composition of the SAM covering the surface. This

characteristic makes it possible to use organic synthesis to tailor organic and

organometallic structures at the surface with positional control approaching ~0.1

nm. SAMs can be fabricated into patterns having 10-100 nm scale dimensions

parallel to the surface. SAMs are well-suited for studies in nanoscience and

technology because of the following characteristics:

1) They are easy to prepare, that is, they do not require ultrahigh vacuum

(UHV) or other specialized equipment (e.g. Langmuir-Blodgett (LB) troughs) in

their preparation.

2) They form on objects of all sizes and are critical components for

stabilizing and adding function to preformed, nanometer-scale objects (for

example, thin films, nanowires, colloids, and other nanostructures, they are also

suitable for biomolecules immobilization).

3) They can couple the external environment to the electronic (current-

voltage responses, electrochemistry) and optical (local refractive index, surface

plasmon frequency) properties of metallic structures.

4) They link molecular-level structures to macroscopic interfacial

phenomena, such as wetting, adhesion, and friction.

23

5) Flexibility to design the head group of SAM with various functional

groups in order to accomplish hydrophobic or hydrophilic surface as per the

requirements.

6) Ability to unravel molecular level information about phenomena such as

protein adsorption, DNA hybridization, antigen-antibody interaction etc. using

surface sensitive techniques such as scanning probe microscopies.5

1.2.3 Types of Substrates and Characterization of SAMs

The substrate is the surface on which a SAM forms. Types of substrates

range from planar surfaces (glass or silicon labs supporting thin films of metal,

metal foils, single crystals) to highly curved nanostructures (colloids,

nanocrystals, nanorods). Planar substrates are widely used for characterizing the

structure-property relationships of SAMs because they are easy to prepare and

compatible with a number of techniques for surface analysis and physicochemical

characterization such as cyclic voltammetry, scanning probe microscopies,

infrared reflectance absorption spectroscopy (IRRAS), Raman spectroscopy, X-

ray photoelectron spectroscopy (XPS), near edge X-ray absorption fine structure

spectroscopy, contact angle goniometry, optical ellipsometry, surface plasmon

resonance spectroscopy, and mass spectrometry. Other metallic nanostructures,

such as nanoparticles, can also support SAMs, and these systems can be

characterized by many other techniques.

The structures of SAMs and the mechanisms by which they assemble are

topics that have evolved considerably over the past two decades, particularly

because there have been substantial advances made in methods suitable for

characterizing them. The development of scanning probe microscopies provided

powerful new capacities to study both the structural organization of SAMs and the

assembly process at a molecular level. These techniques have greatly extended the

initial structural understandings derived mainly from spectroscopic techniques

(IRRAS, XPS, ellipsometry, etc.) and physical methods (principally studies of

wetting). The extensive literature on SAMs has established a common point of

view that SAMs naturally exhibit a high degree of structural order after assembly

and, therefore, form well-defined phases of organic groups organized in precisely

24

understood lateral organizations on the underlying substrate. In fact, SAMs are

dynamic materials that include significant forms of structural complexities,

especially when immersed in fluids. As a rule, SAMs embed intrinsic and

extrinsic defects because they adopt adsorbed structures that are directed by the

thermodynamics of a reasonably complex chemisorption process.

Figure 5. Schematic illustration of some of the intrinsic and extrinsic defects

found in SAMs formed on polycrystalline substrates.

The cartoon of Figure 5 shows how SAMs can be substantially more

complex than the highly ordered arrangements that are commonly assumed. The

causes of defects in SAMs are both intrinsic and extrinsic. Examples of external

factors include cleanliness of the substrate, the methods for preparing the substrate

and the purity of the solution of adsorbate. On the other hand, these defects can be

eliminated by proper control of the experimental conditions. Examples of intrinsic

factors are the many structural defects that the substrate has itself and the complex

phase behaviors due to the dynamic system of the SAMs. These defects can be

minimized but never completely eliminated.

The assembly process involves a thermodynamic equilibrium between

adsorbates on the surface and their precursors free in solution. Although these

SAMs may be kinetically stable in the absence of a flux of adsorbate, the high

coverage of the adsorbate present in the SAM is, in fact, thermodynamically

unstable. Only in a case where the rate of desorption is rigorously zero would the

SAM be expected to exist for an indeterminate period outside the solution used to

prepare it.

25

1.2.4 Nature of the Metal-SAM Interface

Most SAMs of practical interest are formed at reactive interfaces. The

adsorbate and the substrate are both transformed to some degree by the reactions

that lead to the formation of the SAM itself. The chemistry involved for the

chemisorption of thiols on gold is in principle the most straightforward, but it

probably remains as the most enigmatic. Because gold does not form a surface

oxide (as, for example, does silver), the formation of SAMs from thiols is not

complicated by chemistry that might be required to displace or reduce surface

oxides. On the other hand the details regarding the nature of the metal-sulfur bond

and the spatial arrangement of the sulfur groups on the underlying gold lattice are

still controversial.

The formation of a thiolate requires the chemical activation of the S-H

bond of the thiol (or the S-S bond of the disulfide). It is established that the

adsorption of dimethyl disulfide on Au(111) occurs dissociatively.57 The reaction

is reversible, and recombinative desorption of the disulfide is an activated process

with a barrier of ca. 30 kcal/mol. This energy suggests that a fairly significant

degree of charge transfer to sulfur must occur in the thiolates.58 Of particular

interest was the estimation that the barrier for the bimolecular recombinative

desorption of an alkanethiolate from a SAM on gold in the form of a dialkyl

disulfide is ~15 kcal/mol.59 This value is approximately a factor of 2 less than that

deduced in the gas-phase studies. We note here, though, that the two energies are

not directly comparable given that one also contains contributions from the heats

of dissolution of the adsorbate as well as the heat of immersion of the substrate in

the solvent. In this context, the range of reported values appears to be one that

follows directly from the different forms of the measurements used to assess the

strength of the Au-S bonding interaction. As the vacuum measurements are most

easily interpreted, it is reasonable to conclude that the Au-S bond that anchors the

SAM is, in fact, a reasonably strong one (a homolytic Au-S bond strength on the

order of ca. -50 kcal/mol) based on the known S-S homolytic bond strength of a

typical dialkyl disulfide (~62 kcal/mol).60

The fate of the hydrogen of the S-H groups still has not been determined

unambiguously. It seems probable that adsorption in a vacuum leads to loss of the

26

hydrogen in the form of molecular hydrogen. The reductive elimination of H2

from Au(111) is a weakly activated process. In aqueous solution, another

possibility exists. If the thiol hydrogen is not lost in the form of H2, the presence

of oxygen in the reaction medium might also lead to its oxidative conversion to

water. In either case, the Au-S bonding interaction in the thiolate is sufficient to

retain the chains at the surface in a durable fashion and preclude a recombinative

desorption of a disulfide product at room temperature.

The central bonding habit of the high-coverage alkanethiol phases on

Au(111) is generally accepted to be based on a (√3x√3)R30° overlayer (R=

rotated).61-75 Figure 6 shows this structure schematically. The SAMs formed by n-

alkanethiols are usually described as simple thiolate adlayers (chemisorbed

structures formed by the activation of the S-H bond at the gold surface).66-67

Within this model there has been considerable discussion of the surface sites

involved in this bonding. Most studies of SAMs on gold have employed substrates

presenting a strong (111) texture to support the monolayer. Other studies have

been directed at different crystallographic textures, although the structural

literature available in these cases is far more limited.65,68

Figure 6. Schematic diagram depicting the arrangement of decanethiolates on

Au(111) lattice when maximum coverage of the thiolates is attained. (a) The

arrangement shows a (√3x√3)R30° structure where the sulfur atoms (dark gray

circles) are positioned in the 3-fold hollows of the gold lattice (white circles). The

light gray circles with the dashed lines indicate the approximate projected surface

area occupied by each alkane chain; the dark wedges indicate the projection of

the CCC plane of the alkane chain onto the surface. The alkane chains tilt in the

direction of their next-nearest neighbors; (b) Cross-section of the SAM.

27

1.2.5 Organization of the Organic Layer

The geometric arrangement of the sulfur moieties on the surface and the

nearest-neighbor distances between the metal atoms at the surface are factors that

determine the upper limit on the density of molecules on the surface. This two-

dimensional density of molecules may not correspond, however, to the density

that the same molecules could attain in a crystalline form. The arrangement of

molecules that is dictated by the placement of the sulfur moieties on the surface

may not maximize the lateral interactions between the organic components of the

SAMs. To minimize the free energy of the organic layer, the molecules adopt

conformations that allow high degrees of Van der Waals interactions and, for

some molecules such as peptides, hydrogen bonds with the neighboring

molecules; these arrangements yield a secondary level of organization in the

monolayer that is important in determining macroscopic materials properties, such

as wetting and conductivity of the SAMs.

Figure 7. a) Schematic view of an all-trans conformer of a single, long-chain

alkanethiolate adsorbed on a surface. The tilt angle (α) is defined with respect to

the surface normal direction. The twist angle (β) describes the rotation of the

CCC bond plane relative to the plane of the surface normal and the tilted chain;

b) Schematic views of single, long-chain alkanethiolates (with even and odd

numbers of methylene groups) adsorbed on gold. The conserved value of α for

each produces different projections of the terminal methyl group on the surface.

28

A simple single-chain model is sufficient to facilitate comparisons of the

organization adopted by different organosolfur compounds with (mostly) linear

conformations on different types of substrates (Figure 7a). Two parameters

describe the variations in the orientation of the organic molecules in the SAM: the

angle of tilt for the linear backbone of the molecule away from the surface normal

(α) and the angle of rotation about the long axis of the molecule (β). As defined in

Figure 7, α can assume both positive and negative values; values of β range from

0° to 90°. For SAMs formed from n-alkanethiols on gold, palladium, silver,

copper, mercury, platinum, and other metals, the alkane chains adopt a quasi-

crystalline structure where the chains are fully extended in a nearly all-trans

conformation. The tilts of these chains vary for the various metals: the largest

cants (α, with an absolute value near 30°) are found on gold, while the structures

most highly oriented along the surface normal direction arise on silver (α ~10°)

and mercury (α ~0°). The average β for gold lies near 50°, while for other metals,

the data, where available, indicates values generally clustered near 45°. These data

are consistent with space-filling models involving (at least for the case of gold)

chain tilts lying along the direction of the next-nearest neighbor, i.e., an ordered

structure involving a hexagonal arrangement of the sulfur atoms. These

assumptions have been confirmed by the results of diffraction studies.69,70

Not all thiolated molecules adopt the same orientations as n-alkanethiols.

For cases where the steric requirements of the adsorbate preclude the ordering

found for the n-alkanethiolate structures, evidence of other organizations has been

detected. The values of α for SAMs formed by n-alkanethiols on Au(111) appear

to be unique. The tilt of the chain projects an orientation of the average chain in

which the sign of the tilt angle is conserved regardless of the number of carbons in

the alkane chain. All available data suggest that the structures exhibited by

thiolate SAMs on gold adopt a value of α ~ +30°. This feature of the assembly

leads to very different surface projections of the methyl groups for SAMs with

odd and even numbers of methylene groups (Figure 7b) and correlates strongly

with the unique wetting behaviors of SAMs on gold; SAMs of thiolates with an

odd number of methylene groups produce surfaces whose free energies are

systematically slightly larger than those with an even number of methylenes.

29

1.3 Gold in Electrochemical Biosensors

There are five characteristics of gold that make it a good choice as a

substrate for studying SAMs. First, gold is easy to obtain, both as a thin film and

as a colloid. It is straightforward to prepare thin films of gold by physical vapor

deposition, sputtering, or electrodeposition. Although expensive and not essential

to most studies of SAMs, single crystals are available commercially. Second, gold

is exceptionally easy to pattern by a combination of lithographic tools

(photolithography, micromachining, etc.) and chemical etchants. Third, gold is a

reasonably inert metal: it does not oxidize at temperatures below its melting point;

it does not react with atmospheric O2 and it does not react with most chemicals.

These properties make it possible to handle and manipulate samples under

atmospheric conditions instead of under UHV. Gold binds thiols with high

affinity, and it does not undergo any unusual reactions with them. Four, thin films

of gold are common substrates used for a number of existing spectroscopies and

other analytical techniques. This characteristic is particularly useful for

applications of SAMs as interfaces for studies in biology. Five, gold is compatible

with cells. SAMs formed from thiols on gold are stable for periods of days to

weeks when in contact with the complex liquid media required for cell studies.

The properties of gold nanoparticles (AuNPs), such as light absorption and

their excellent electroactivity, are bringing interesting immunosensing

alternatives. Particular emphasis is given to the different optical71 and

electrochemical72,73 detection methodologies where NPs show significant impact.

In certain cases, assays based on nanomaterials have offered significant

advantages over conventional diagnostic systems with regard to assay sensitivity,

selectivity, and practicatility.74 AuNPs are so small that they exhibit

characteristics that are often not observed in the bulk materials. This is due to the

quantum size effect that leads to unique optical, electronic, and catalytic

properties.75 AuNPs are also fully compatible with biomolecules, when decorated

with thin organic coatings. This has resulted in their use in sensors for DNA,

proteins, organic analytes, and metal ion. The use of thiol groups for their

functionalization is a good way to control the direction of the bond between the

30

label and the biomolecule. NPs’ involvement in DNA, protein and even cell

sensing systems, have recently been the most important topics in

nanobiotechnology.75

The electroactivity of AuNPs allows the use of both electrical and

electrochemical techniques for their detection, which allowed to detect low

concentrations of proteins.76 NPs can be directly detected due to their own redox

properties or indirectly due to their electrocatalytic properties toward other

species, such as silver ion reduction.

It is also worth mentioning that their large surface coupled with an easy

bioconjugation make NPs excellent carriers of other electroactive labels in

immunoassay.75 Nanoscale structures of AuNPs on conductive surface combined

with high electrical conductivity can facilitate fast ET to and from redox enzymes,

for examples horseradish peroxidase, providing a sensitive platform for

biosensors. AuNPs have been employed as nanoelectrode relay units transporting

electrons efficiently and activating enzyme bioelectrocatalysis.9 The introduction

of NPs into the traducing platform is commonly achieved by their adsorption onto

conventional electrode surfaces in various forms, including that of a composite.75

To summarize, modified AuNPs electrodes have very large surface areas,

are simple to fabricate and functionalized, retain metallic conductivity, and have

facile biomolecule attachment.9

1.4 α-Aminoisobutyric Acid 310-Helices as

Adsorbate for SAMs

The nature, stability and, for electrochemical biosensors, the electron

transfer properties of the SAM are of paramount importance to determine the

performance of a device or sensor. A question now arises: which are the

molecules of choice to make robust and performing SAM-based electrochemical

devices? SAMs should be sufficiently chemically and electrochemically stable,

well organized, and based on easily tunable molecules. The latter aspect includes

the ease by which the length (which implies modulating the SAM thickness) and

functionalization (with suitable groups on the molecule end facing the solution) of

31

the thiolated molecule can be controlled. The presence of further specific

structural features is also useful, such as the presence of intermolecular hydrogen

bonds between adsorbate molecules (such as C=O⋅⋅⋅H–N hydrogen-bonds

between embedded amide groups) which increases the SAM robustness. The

molecules that conveniently collect all these features, including their particular

compatibility with biomolecules, are peptides. The following question now is:

among all possible systems, which peptides are likely to provide particularly

suitable systems?

In this Thesis research, we also describe some self-assembly features of

thiolated α-aminoisobutyric (Aib) acid homooligomers on activated Au surfaces.

There are several reasons for considering these peptides as good candidates. Aib

is characterized by marked hindrance at the α-carbon and restricted torsional

freedom (Figure 8).77,78

H2N

O

Figure 8. α-Aminoisobutyric Acid.

Owing to these features and differently from peptides based on coded α-

amino acids which form stable helices only for rather long oligomers,79

Aib peptides adopt a 310-helical structure and are rigid even when short.80

Rigidity is ensured by a strong framework of intramolecular C=O···H-N hydrogen

bonds that causes C=O and N-H groups to align significantly along the peptide

axis81 and, therefore, a strong oriented dipole moment arises.

In 310-helices, each intramolecular C=O···H-N hydrogen bond involves

residues i and i + 3, a single helical turn requires 3.24 amino acid residues, the

peptide length increases by 1.94 Å/residue and thus addition of a single coil

increases the peptide length by 6.29 Å.78 The 310-helix is thus more elongated and

thinner than the α-helix, which typically involves 3.63 residue/coil and a vertical

pitch of 1.56 Å/residue. In Figure 9 and in Table 1 we compare different

properties of the two helices. As opposed to simple alkanethiols where an increase

in the number of units increases molecular flexibility, Aib peptides become even

32

stiffer as the number of residues, and thus of intramolecular hydrogen bonds,

increases. 310-helices become more tightly wound as they become longer due to

the shortening of the H-bonds that accompanies the stronger cooperative

interactions.81c

Figure 9. Section of 310-helix (on the left) and section of α-helix (on the right).

Table 1: Average parameters for right-handed 310 and α-helices.

Parameters:

310-Helix α-Helix

Φ 57° 63°

Ψ 30° 42°

N ···O=C H-bond angle 128° 156°

Rotation (per residue) 111° 99°

Axial translation (per residue) 1.94Å 1.56Å

Residues per turn 3.24 3.63

Pitch 6.29Å 5.67Å

In addition and as opposed to simple alkanethiols for which the increase in

the number of methylene groups increases the flexibility of ligands, the key motif

of Aib peptides is the increase of the peptide stiffness with the number of residues

and thus of hydrogen bonds. This feature and the unique rigidity of short Aib

homopeptides are being successfully exploited as spacers or templates in

electrochemical and spectroscopic investigations, also because of the excellent

electronic communication provided by the peculiar backbone stiffness. Since 2D

(and 3D SAM, such as in monolayer protected clusters (MPCs)) formation already

33

causes the conformational freedom of alkanethiols to decrease dramatically, Aib

peptides are expected to behave as even more rigid adsorbates.

The 310-helix helical structure is stable both in the solid state and in

solution.78,80,82 It is also kept unaltered when thiolated Aib peptides are self-

assembled on 1-2 nm gold nanoclusters.83 With nanoparticles, compelling

evidence showed that these peptides also form interchain hydrogen bonds,

resulting in the formation of strong molecular networks. Insights into this

experimental observation were obtained by a multilevel molecular modeling

study.84 A previous analysis of the amide I and amide II regions of IRRAS spectra

of thiolated Aib hexapeptide SAMs on extended gold surfaces also pointed to the

presence of helices.85

The Maran group has previously studied ET reactions across Aib homo-

oligopeptides using donor-peptide-acceptor systems in solution.86 Evidence was

obtained for a mild distance dependence of the ET rate and even an increase of the

rate at a certain peptide length. This outcome was rationalized by considering that

while addition of a new α-amino acid unit increases the donor-acceptor distance,

it also introduces new intramolecular hydrogen bonds that act as efficient ET

shortcuts, thereby counteracting the usually observed exponential drop of the ET

rate with distance. Theoretical studies supported the general features of this

experimental finding.87 These results suggested us that Aib peptides may indeed

furnish ideal molecular bridges for making robust SAMs (molecular rigidity,

interchain network) for electrochemical biosensoring and other

nanobiotechnological applications. Based on the outcome of the solution studies,

efficient electron conduction through Aib-peptide SAMs was conceivably

expected, with useful consequences for detection of analytes through observation

of high redox currents when the SAM is the central part of a transducing

electrochemical platform. Very recently some thiolated Aib-peptides on mercury

electrodes and found that tight SAMs form.88 Although the main target of that

study was to estimate the surface dipole potential of the Aib-peptide SAMs, it was

also observed that the kinetics of the voltammetric reduction of Eu(III) was not

severely slowed by the presence of the peptide film: this also concurred to suggest

that Aib chains may mediate electron tunneling very efficiently. For peptides

thiolated on the nitrogen terminus, where the negative pole of the peptide dipole is

34

located, it was observed that the negatively polarized electrode produces an

interfacial electric field liable of orienting the peptide dipole even against its

“natural” dipole moment.

Very recently, a study coupled to this Thesis work specifically addressed

the issue of determining the ET rate-constant falloff with distance using a series of

Aib-peptide SAMs on gold surfaces.89 The peptides were devised to give raise

from zero to five C=O•••H-N intramolecular hydrogen bonds. The peptides were

thiolated on the positive end of the molecular dipole. The standard heterogeneous

ET rate constants for the chemically-reversible reduction of a soluble redox probe,

Ru(NH3)6Cl3, were determined by CV in 0.5 M KCl aqueous solution. It was

shown that once experimental procedures and surface packing are controlled it is

indeed possible to carry out accurate studies of the distance dependence of ET

through SAMs even when using soluble redox probes. Figure 10 shows the

distance dependence of the ET rate. The values show that the ET rate constant

initially decreases as the peptide length increases but then displays a remarkably

shallow dependence for sufficiently long peptides, the exponential-decay factor

(from 3+ to 5+) being of only 0.08 Å-1. These results show that Aib peptides are

indeed very good mediators of electron tunneling and thus emerge as remarkably

good candidates to make SAMs for electrochemical biosensors.

Figure 10. Calculated values of k° against the number of hydrogen bonds.

H-Bonds

0+ 1+ 2+ 3+ 4+ 5+

ln k

°

-5.0

-4.5

-4.0

-3.5

-3.0

-2.5

-2.0

35

1.4.1 Characteristics of the Investigated Aib-

Homopeptides

Two series of peptides were synthesized, differing in the direction of the

dipole moment: a plus series with the positive pole of the dipole moment oriented

toward sulfur and a minus series where the dipole moment is reversed. The name

of each peptide is given as composed by a number, which refers to the number of

intramolecular hydrogen bonds, and a plus or minus sign, which refers to the

orientation of the dipole moment.

Figure 11 shows the primary and secondary structure of peptide 5+. This

peptide is used in the following discussion to explain the general design of our

systems. The head of our system is represented by the –SH. This group allows the

peptide to be attached to the gold surface. Between sulfur and the α-amino acid

chain there is a -CH2-CH2- spacer meant to facilitate the coordination to the

surface, which otherwise would be hindered by the strong steric hindrance of the

310-helix. On the other terminus, there is a t-butyl group. Figures 11a and 11b help

to understand better the orientation of the dipole. The helicity of the system infers

a particular orientation to the Aib residues, orienting all C=O groups toward the C

terminus and all N-H groups toward sulfur. Accordingly, the dipole moment (in

the plus series) has the plus pole by the thiol group. Table 2 shows the primary

structures for all the peptides belonging to the plus series.

For the minus series three peptides were synthesized (1-, 3- and 5-), to

investigate the effect of reversing the dipole moment and to compare the results

with the plus series, which is our reference series. In the minus series the

characteristics of the design are the same, but while the SH group is now attached

to the C terminus that now becomes the head of the peptide, the t-butyl group is

on the other terminus that becomes the tail. Figure 12 shows the primary (a) and

secondary structure (b) of 5-. From the 3D structure we can notice that the C=O

groups are now oriented toward SH, while the NH groups are oriented toward the

tail of the peptide. Consequently, the negative pole of the dipole moment is by the

thiol group. Table 3 shows the primary structures for all the peptides belonging to

the minus series.

36

a)

b)

Figure 11. a) Primary structure of the peptide. b) Secondary structure of the

peptide.

Table 2: Primary structure of the investigated peptides for the plus series.

Plus Series

Primary structure Name

SH

O

NH

O

NH

0+

NH

SH

O

NH

O

NH

O

1+

SH

O

NH

O

NH

O

NH

O

NH

2+

NH

SH

O

NH

O

NH

O

NH

O

NH

O

3+

SH

O

NH

O

NH

O

NH

O

NH

O

NH

O

NH

4+

SH

O

NH

O

NH

O

NH

O

NH

O

NH

O

NH

NH

O

5+

SH

O

NH

O

NH

O

NH

O

NH

O

NH

O

NH

NH

O

37

a)

NH

O

NH

O

NH

O

NH

O

NH

O

NH

NH

OO

SH

b)

Figure 12. a) Primary structure of the peptide. b) Secondary structure of the

peptide.

Table 3: Primary structure of the investigated peptides for the minus series.

Minus Series

Primary structure Name

NH

SHNH

O

NH

OO

1-

NH

SHNH

O

NH

O

NH

O

NH

OO

3-

NH

O

NH

O

NH

O

NH

O

NH

O

NH

NH

OO

SH

5-

38

1.5 Purposes of the Thesis

This Thesis concerns the development of a very efficient immunosensor

electrochemical device for detecting cancer biomarker proteins, and possible ways

to improve its efficiency by modifying parts of the system that are relevant to

improve stability and ET through the modified electrode surface.

For the part concerning the study of the ultrasensitive electrochemical

immunosensor (Figure 13) for early cancer biomarkers detection, I worked with

the research group of Prof. James F. Rusling (Department of Chemistry,

University of Connecticut, Storrs, CT, USA). The research associated with the

study of the stability and structure of Aib-peptide self-assembled monolayers to

be used for improving the sensor performance was carried out in the group of

Prof. Flavio Maran. The two topics are related because these peptides will be used

to anchor the antibody onto AuNPs. This, however, is an aspect that for time

limitations could not be covered during my Thesis work. Details on the Thesis

scope and structure are as follows.

We focused our attention on Nanog detection. Nanog is a protein that may

be involved in carcinogenesis of cervix and progression of cervical carcinoma.

Nowadays, the researchers still do not know the detection limit of this biomarker

and the difference of concentration between healthy individuals and patients with

cancer. Therefore, we aimed at making an electrochemical sensor capable of

displaying very high-sensitivity immunoarrays and low detection limit. Sensors

were prepared and, particularly, several conditions to make Nanog-based

electrodes were essayed. Eventually, we could optimize the conditions and obtain

a nice calibration plot (Amperometric current versus Nanog concentration). Most

of the initial work was carried out using the sample handling technology, but then

we integrated the system into a microfluidic device, the goal being to automate the

method as much as possible.

To improve the efficiency, we are about to further optimize the

immunosensor by changing some elements of the transducer, particularly by using

a SAM formed by peptides allowing very fast ET (see below) and by increasing

the active superficial area thanks to nanostructured gold electrodes

alternative to a bed of

Figure 13. AuNP immunosensor with Ab

from a sample after treating with Ab

enzyme labels for each Nanog. The detection step involves immersing the sensor

into buffer containing mediator, applying voltage, and injecting H

We carried out an investigation of related issues by using SAMs formed

with thiolated Aib peptides of different lengths. The effect of the orientation of the

peptide dipole moment was studied by a

nitrogen or carbon terminus. The stability and conformational properties of such

SAMs were assessed by

the IR absorption spectroscopy of the free peptides

these SAMs Aib peptides form 3

bonds, and pack tightly, the surface coverage depending on both the peptide

length and orientation. We also found that short peptides may undergo helix

disruption, with formation of structures where the n umber of inter

interactions increases. The results nicely support

concerning the chemical and electrochemical stability of these SAMs as well as

the efficiency of ET through t

know which peptides should provide the best transducer substrate supporting the

actual Nanog-sensor architecture.

the active superficial area thanks to nanostructured gold electrodes

to a bed of AuNPs.

AuNP immunosensor with Ab1 attached that has captured an antigen

from a sample after treating with Ab2-magnetic-bead-HRP providing multiple


containing mediator, applying voltage, and injecting H



peptide dipole moment was studied by attaching the thiolated moiety to either the


SAMs were assessed by an extensive IRRAS investigation, in comparison with

the IR absorption spectroscopy of the free peptides. This study showed that in

these SAMs Aib peptides form 310-helices, form interchain C=O••



disruption, with formation of structures where the n umber of inter

interactions increases. The results nicely supported what recently found

chemical and electrochemical stability of these SAMs as well as

the efficiency of ET through them. Main outcome of this study is that we now


sensor architecture.

39

the active superficial area thanks to nanostructured gold electrodes as an

attached that has captured an antigen

HRP providing multiple


containing mediator, applying voltage, and injecting H2O2.



ttaching the thiolated moiety to either the


investigation, in comparison with

s study showed that in

••H-N hydrogen



disruption, with formation of structures where the n umber of inter-chain

what recently found

chemical and electrochemical stability of these SAMs as well as

Main outcome of this study is that we now


40

2. Experimental

2.1 Chemicals

Immunosensor. Polydiallyldimethylammonium chloride (PDDA), L-

gluthathione reduced (99%), gold (III) chloride trihydrate (99.9%), 2,2’-azino-

bis(3-ehtylbenzthiazoline-6-sulfonic acid), horseradish peroxidase (HRP, MW

44000), 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC), N-

hydroxysulfosuccinimide ester (NHSS) and Lyophilized 99% bovine serum

albumin (BSA), and Ru(NH3)6Cl3 (98%) were purchased from Sigma-Aldrich.

The primary antibody (Ab1), the secondary antibody (Biotin-Ab2), and the Nanog

antigen were obtained from Dr. Dharamainder Chaudary UCONN Health Center

(Farmington, CT). Immunoreagents were dissolved in a pH 7.0 phosphate saline

(PBS) buffer (0.01 M in phosphate, 0.14M NaCl, 2.7mM KCl). Carboxyl

functionalized magnetic beads were obtain from Polysciences, Inc. The

poly(dimethoxy)silane (PDMS) kit was from Dow Corning. Hydroquinone (HQ)

and hydrogen peroxidase (H2O2, 30%) were from Fisher. Water was deionized.

SAMs. Ethanol (HPLC grade, ≥99.8%), hydrogen peroxide 30%, and

sodium hydroxide monohydrate (≥99.9995%) were purchased from Fluka.

Sulfuric acid 98% (Aristar Grade) was purchased from BDH. Low conductivity

water was Water pro analysis obtained from Merck.

2.2 Preparation of HRP single electrodes

The steps to fabricate the immunosensor platform on a single electrode are:

1. Absorption of a layer of cationic polydiallyldimethylammonium

(PDDA) from aqueous solution onto a pyrolytic graphite (PG) disk electrode. We

used a 20 µL solution of PDDA, for 25 min.

41

2. The electrode was washed with H2O and the water removed by shaking

the electrodes.

3. 20 µL of a diluted (1:5) solution of 2 mg/mL glutathione-decorated gold

nanoparticles [GSH-AuNP] in HPES buffer (pH = 8), were poured onto the

PDDA-modified electrode, utilizing the standard layer-by-layer alternate

electrostatic adsorption approach90 to form a dense layer with organic

functionality for capture antibody attachment. Incubation required 25 min.

4. The sensor was washed with H2O.

5. To activate the carboxylic acid groups of the gold platform, an aqueous

solution of 400 mM EDC [1-ethyl-3-(3-dimethylaminopropyl) carbodiimide] and

100 mM NHSS [N-hydroxysulfosuccinimide ester] was prepared. The electrode

was covered with 30 µL of this solution.

6. After 10 min, the excess reagent was removed by washing the electrode

with H2O.

7. The EDC-activated particles were reacted with amine groups of the

primary antibody Ab1. It is known5 that immobilization of Ab1 does not diminish

its bioactivity. For this step, 20 µL of a 25 µg/mL solution was drop casted on the

sensor. The electrode, covered with a wet Vial, was then put in an oven at 37 °C

for 2.5 h.

8. To remove all not-binding elements, the electrode was washed for 3 min

with PBS buffer with 0.05% Tween-20 surfactant, and then for 3 min with PBS

buffer (0.01M pH=7.0). The electrode was shaken to remove excess water buffer.

42

9. The inhibition of nonspecific binding of labeled detection antibodies is

crucial to achieve high sensitivity and low detection limits. To optimize this step,

the surface was treated with 20 µL of a 0.2% BSA solution [Bovine Serum

Albumin] in PBS. The electrode was then put in an oven at 37 °C for 45 min.

10. Step 8 was repeated.

11. 10 µL of the given test solution of Nanog Antigen (NA) were

incubated onto the surface. We used various concentrations; for example, if the

procedure involved activation of 9 electrodes, three were used for the control (0

pg/mL NA), three for the first concentration (e.g., 25 pg/mL NA), and three for a

higher concentration (e.g., 50 pg/mL NA). The electrodes were put in an oven at

37 °C for 1 h.


13. The sensors with bound Nanog were then incubated using 10 µL of a

1µg/mL solution of the secondary antibody Biotin-Ab2. The electrodes were put in

an oven at 37 °C for 45 min.


43

15. The final incubation step consisted in labeling Ab2 with 10 µL of a

5µg/mL solution of Srt-HRP (horseradish peroxidase enzyme, functionalized with

Streptavidin). The electrodes were put in an oven at 37 °C for 30 min.

16. The electrodes were washed for 3 min with PBS-T20 buffer, and then

for 3 min with PBS buffer (0.01M pH = 7.0). The electrodes were covered with

wet Vials and put in a refrigerator.

To amplify sensitivity and improve the detection limit, a different

procedure was also applied, starting from step 13.

13’. In a separated eppendorf we prepared 100 µL of a solution composed

by 0.2 % BSA PBS buffer (pH = 7.0) solutions: 50 µL of 5 µg/mL solution of

multiple HRP labels; 30 µL of 2.5 µg/mL solution of magnetic beads (FeO2, with

a diameter of ∼ 1 µ); 20 µL of 2500 µg/mL solution of Ab2. 10 µL of a dilute

solution of this mixture was used for incubating the electrodes.

14’ The electrodes were washed for 3 min with PBS-T20 buffer, and then

for 3 min with PBS buffer (0.01M pH = 7.0). The electrodes were covered with

wet Vials and put in a refrigerator.

.

44

2.3 Fabrication of the Microfluidic D

and Preparation of Eight

To prepare the microfluidic channel

base and curing agent were mixed in 10:1 ratio, stirred vigorously for 5 min, and

then degassed for 30 min under dynamic vacuum to remove all air bubbles. The

clear solution was poured onto a negative mold and heated at 85°C

After cooling, the linear

placed between two flat and machined poly(methylmethacrylate) (PMMA) plates

to provide a microfluidic channel (Figure 14).

a)

b)

Figure 14. a) Photographs of the PMMA device and

photograph of the entire microfluidic device; c) photograph of the molds for the

PDMS channel.

The channel is ~1 mm wide

volume. The top PMMA plate was mac

equipped with female ports (4

Fabrication of the Microfluidic Device

Preparation of Eight-Electrode

To prepare the microfluidic channel, the PDMS kit was used:



clear solution was poured onto a negative mold and heated at 85°C for 2.5

, the linear-shape PDMS was pealed off the mold, then


to provide a microfluidic channel (Figure 14).14

c)

a) Photographs of the PMMA device and the PDMS channel; b)


The channel is ~1 mm wide, ~2 mm thick, 27 mm long and carries ~54

volume. The top PMMA plate was machined with one inlets and one outle

equipped with female ports (4 mm diameter) for screwing in standard plastic

evice

Array

the PDMS



for 2.5 h.14

shape PDMS was pealed off the mold, then


the PDMS channel; b)


hick, 27 mm long and carries ~54 µL

hined with one inlets and one outlet and

mm diameter) for screwing in standard plastic

fittings (1.5 mm i.d., Upchurch) to hold connecting 0.2

Sample were injected by a syringe pump (Harvard, no.

inlet via an injector valve (Rheodyne, no.

15). The top PMMA substrate is equipped with four holes (0.5 and 0.2

diameter) directly above the microfluidic channel; two for inserting Ag/AgCl wire

used as a reference and two for inserting Pt wire as counter electrode.

Figure 15

For the eight-electrode array system (

manufactured by Kanichi Research Ltd, Mancester, UK), we essentially followed

the same functionalization procedure used for the single rotati

(section 2.2). The difference was that while the PDDA,

put manually, the BSA 0.2%, the Nanog

injected by using the syringe pump. For this purpose

pump was filled up with the PBS buffer and the flow

~50 s after injection of the BSA solution with the injection valve, the fluid

could reach the channel and cover completely all electrodes.

was blocked to permit th

bubble, otherwise the sensor is compromised.

and Ab2-MP-HRP. After every incubation, the sensor was washed with PBS and

PBS-T20.

mm i.d., Upchurch) to hold connecting 0.2 mm i.d. tubing (PEEK).

Sample were injected by a syringe pump (Harvard, no. 55-3333) connected to one

ector valve (Rheodyne, no. 9725i) via 0.2 mm i.d. tubing (Figure

). The top PMMA substrate is equipped with four holes (0.5 and 0.2


used as a reference and two for inserting Pt wire as counter electrode.

Figure 15. Photograph of the entire microfluidic system.

electrode array system (arrays of eight graphite electrodes were

by Kanichi Research Ltd, Mancester, UK), we essentially followed

the same functionalization procedure used for the single rotating disk electrode

). The difference was that while the PDDA, the AuNPs and Ab

the BSA 0.2%, the Nanog and the Ab2-MP-HRP

injected by using the syringe pump. For this purpose, the syringe of the Harvard

pump was filled up with the PBS buffer and the flow rate set to 50

s after injection of the BSA solution with the injection valve, the fluid

could reach the channel and cover completely all electrodes. At this point the flow

was blocked to permit the incubation. For this step is very important

ise the sensor is compromised. The procedure was

HRP. After every incubation, the sensor was washed with PBS and

45

mm i.d. tubing (PEEK).

3333) connected to one

i.d. tubing (Figure

). The top PMMA substrate is equipped with four holes (0.5 and 0.2 mm


used as a reference and two for inserting Pt wire as counter electrode. 14

Photograph of the entire microfluidic system.

arrays of eight graphite electrodes were

by Kanichi Research Ltd, Mancester, UK), we essentially followed

ng disk electrode

the AuNPs and Ab1 were

HRP mixture were

the syringe of the Harvard

50 µL/min.

s after injection of the BSA solution with the injection valve, the fluid

At this point the flow

important to avoid any

repeated for NA

HRP. After every incubation, the sensor was washed with PBS and

46

2.4 Gold Preparation and SAM formation

Formation of reproducible and well-packed Aib-based peptide SAMs is

very sensitive to the cleaning – activation procedure. After trying different

procedures, SAM formation was optimized by careful application of the following

protocol leading to very reproducible results. For all experiments we used 11 x 11

mm ArrandeeTM gold plates, which consist of a borosilicate glass substrate

(thickness 0.7 ± 0.1 mm), a thin chromium interlayer (2.5 ± 1.5 nm), and a gold

layer (250 ± 50 nm). We used a new Au plate for each experiment.

The plates were first immersed in ethanol for 5 min, dried in vacuum for

30 min, immersed for 5 min in a freshly prepared but already cold piraña solution

(3:1 v/v solution of concentrated sulfuric acid and hydrogen peroxide. Extreme

caution must be exercised when using piraña solution, as it is a very strong

oxidant and reacts violently with organic matter), and then carefully washed with

a stream of water (deionized and then distilled from potassium permanganate).

After washing, the plates were immersed in water for 5 min, and the procedure

repeated three times. The plates were then dipped into ethanol for 5 min and dried

in a desiccator under vacuum for 30 min. Annealing was carried out using a

butane torch for soldering electronics, in a slightly darkened room. The flame was

set for conical blue (reducing flame) and first allowed to touch obliquely the

borders of the gold surface, and then moved steadily inward in a square fashion:

annealing eventually caused appearance of a dark red glow on the gold surface.

After waiting a few seconds to allow the surface to cool down, the same sequence

step was repeated for an overall 3 min. The plates were left to cool down for 30

min, under a protecting glass cap. The annealing/cooling procedure was repeated

three times. The plates were then immersed in water, ethanol, and finally either

immersed into the peptide-containing ethanol solution or dried under vacuum.

Some experiments were specifically carried out to analyze the condition of

the gold surface. Thus, after annealing and washing, the plates were dried in a

desiccator under argon for 1 h and then either studied by scanning tunneling

microscopy (STM: PicoSPM, Molecular Imaging) or cyclic voltammetry (CV).

STM analysis of the so-prepared plates were carried out at room temperature

47

under ambient conditions using Pt/Ir (80:20, 0.25 mm diameter, Veeco) tips. The

images were recorded in constant current mode and the typical conditions for the

imaging of Au(111) terraces were VT= 200 mV, iT= 100 pA. This technique

revealed that the Au surface displayed a preferred (111) orientation with terraces

as large as ~100 nm that were atomically flat and separated by monoatomic steps

(0.24nm). To further test the outcome of the annealing procedure, we carried out

CV experiments in 0.5 M H2SO4 and potential cycling was performed to include

both the double-layer and the oxide regions of gold. The CV curves largely

showed the pattern expected for the underpotential deposition of oxygen onto a

clean Au electrode, mostly Au(111), in a contaminant-free solution,91 thereby

confirming both the quality and dominant orientation of the annealed surface.

We also calculated the roughness factor of the so-obtained Au plates. The

roughness factor is the ratio between the electrochemical and the geometrical

areas. For these electrochemical experiments, the Au plates were clamped to a

small gold alligator and most of the electrode surface was wrapped with a very

thin Teflon tape. The geometrical areas were calculated by integration, using

pictures of the exposed electrode surface in comparison with area standards. The

determination of the electrochemical area was carried out on the (naked) Au plates

obtained after full reductive desorption of the peptide SAM.89

CV experiments were run at 0.1 – 1 V s-1 in 0.5 M KCl containing

Ru(NH3)6Cl3, whose diffusion coefficient is 6.84 ± 0.06 × 10-6 cm2 s-1.89 The

electrochemical (= active) area could be calculated from the peak current values,

using the equation valid for a reversible ET.92 The average roughness factor was

found to be 1.16 ± 0.08.

After annealing and washing, the freshly prepared gold plates were

immersed for 48 h in a 0.5 mM ethanol solution of the peptide of choice. To

carefully remove any trace of the incubation solution and loose molecules

adsorbed on the monolayer, the gold plates were washed with a stream of ethanol

and then immersed in ethanol for 5 min; these washing steps were repeated three

times. The plates were finally dried under vacuum for 30 min. The

cleaning/activation/incubation procedure was applied just before carrying out the

IRRAS measurements.

48

2.5 Instrumentation and procedures

2.5.1 Electrochemistry

Rotating-Disk Electrode Approach. A CHI 660 electrochemical

workstation was used for cyclic voltammetry and amperometry at ambient

temperature (22 ± 2°C) in a three-electrode cell. Amperometry was carried out at -

0.3 V against the standard saturated calomel electrode (SCE), using a pyrolytic

graphite (PG) working electrode (rotated at 3000 rpm for optimum sensitivity),

and a platinum wire counter-electrode.

Microfluidic Setup. The experiments were performed using a CHI 1010A

eight-channel potentiostat allowing to simultaneously measuring the current

flowing at the eight electrodes inserted in the microfluidic channel. Amperometry

was carried out at -0.2 V vs the Ag/AgCl reference electrode, and using a

platinum wire as the counter-electrode. For cyclic voltammetry experiments, the

flow rate was set a zero, while for the amperometric detection the flow rate was

100 µL/min using 100 µL injector sample loop. Buffer and hydrogen peroxide

solutions were deoxygenized with nitrogen.

2.5.2 IRRAS

FT-IR reflection absorption spectroscopy measurements were carried out

using a Nicolet Nexus Fourier Transform Infrared Spectrometer equipped with a

liquid nitrogen cooled mercury-cadmium-tellurium (MCT) detector. The

measurements were carried out with the grazing angle set at 80°, with a resolution

of 2 cm-1, and each spectrum was the result of an average of 1000 scans. The

optical path was purged with nitrogen before and during the measurements. To

keep the measuring chamber constantly under an inert atmosphere, we set up a

little dry-box containing the entire optical path. The sample and the blank (an

otherwise identical Arrandee gold plate, but with no SAM) were inserted in the

dry-box. Before measurements, the system was purged for 1 h with nitrogen dried

by flowing through a column filled with CaCl2 and silica gel.

49

3. Results and Discussion

3.1 Electrochemical Immunosensors

Electrochemical detection combined with nanoparticle amplification offers

potentially low-cost, high-through put solutions for detection of clinically

significant proteins that have yet to be fully realized. Amperometric sensors, field

effect transistors, and impedance methods are among the approaches being

explored. The sensitivity of an electrochemical sensors can be improved by using

nanostructured electrodes, such as those based on carbon nanotubes or gold

nanoparticles. In the electrochemical detection of proteins on these specific

electrode surfaces by immunoassay protocols, appropriate functional groups on

the nanoparticle facilitate high concentrations of chemically linked capture

antibodies. In this approach, antibodies on the electrode capture analyte proteins

from the sample, the surface is then be treated with an enzyme-labeled secondary

antibody, and the enzyme label is detected electrochemically.

3.1.1 Characterization of AuNP platform

Glutathione-protected gold nanoparticles (GHS-AuNP) were prepared by a

reported method9 and characterized by Transmission Electron Microscopy (TEM),

UV-Vis and Infrared Spectroscopy.

The glutathione protected gold nanoparticle dispersion was analyzed for

core dimensions and size distribution by TEM. Figure 16a shows that all particles

are spherical and are well separated from each other. The size histogram obtained

from analysis of the TEM images revealed that the average diameter of GHS-

AuNP was 5.1 ± 1.4 nm, a relatively narrow size distribution.

Particle size was confirmed by visible absorption spectroscopy band for

the GHS-AuNP dispersion at 508 nm indicating diameter ~ 5 nm (Figure 16c).

50

This absorption peak corresponds to gold plasmon band, confirming that the

solution contained isolated nanometric sized gold particles.

The presence of glutathione on the AuNPs was confirmed by observation

of characteristic carbonyl and amide bands at 1538, 1628, and 1713 cm-1 by IR

Spectroscopy (Figure 16c).

Figure 16. Characterization of glutathione protected gold nanoparticles: a) TEM

image of GHS-AuNPs; b) corresponding size distribution histogram of GHS-

AuNPs showing average size 5.1 ± 1.4 nm; c) UV-Vis absorption spectra of

glutathione protected gold nanoparticle dispersion in HEPES buffer; d) IR

spectra of (1) pure glutathione and (2) glutathione protected gold nanoparticles.

The surface was also characterized by tapping-mode atomic force

microscopy (Figure 17). The initial thin layer of PDDA (0.5 ± 0.2 nm) adsorbed

was relatively smooth with mean surface roughness of 0.13 ± 0.08 nm, almost

twice that of the bare mica. Figure 17 panels b, c shows an AFM image that

changed topography after AuNPs were absorbed onto the PDDA layer. The AFM

data suggest that there is a densely packed nanoparticulate layer, whi

associated with the A

is achieved with a mean surface roughness

the surface valleys and peaks,

increase in surface area with respect to

image obtained after the capture antibody was covalently linked onto the

carboxylated of the AuNP layer using EDC/NHSS chemistry. The densely packed

AuNP layer disappeared and a ro

of any globular protein coated on a

Figure 17. Tapping-mode atomic force microscope images of (a) layer of PDDA

on smooth mica surface; (b) PDDA/AuNP bilayer; (c) phase

PDDA/AuNP bilayer; (d) AuNP platform after covalent linkage of primary

antibodies (Ab1) onto the glutathione carboxylate groups of AuNP.

data suggest that there is a densely packed nanoparticulate layer, whi

d with the AuNPs. A nearly complete coverage on the underlying surface

mean surface roughness, as the arithmetic average deviation of

the surface valleys and peaks, of 1.99 ± 0.11 nm, corresponding to

ncrease in surface area with respect to bare mica. Figure 17 d shows an AFM

after the capture antibody was covalently linked onto the


AuNP layer disappeared and a rolling hill-like appearance, generally characteristic

of any globular protein coated on a rough surface, was observed.9

mode atomic force microscope images of (a) layer of PDDA

on smooth mica surface; (b) PDDA/AuNP bilayer; (c) phase contrast image of


) onto the glutathione carboxylate groups of AuNP.

51

data suggest that there is a densely packed nanoparticulate layer, which can be

nearly complete coverage on the underlying surface

the arithmetic average deviation of

corresponding to about a 24-fold

Figure 17 d shows an AFM

after the capture antibody was covalently linked onto the


generally characteristic

mode atomic force microscope images of (a) layer of PDDA

contrast image of


) onto the glutathione carboxylate groups of AuNP.

52

3.1.2 Electrochemistry of immunosensors

The AuNP/Ab1/Ag/Ab2/MP/HRP electrode, assembled as described in the

Experimental Section, was transferred in an electrochemical cell or in the

microfluidic channel filled with of pH 7.0 buffer and 1mM hydroquinone (HQ). In

both cases, the cyclic voltammetry (CV) was carried out in quiescent solution to

confirm that the electrode was performing well: with the rotating-disk electrode

the angular speed was set to zero and with the microfluidic channel the flow of the

PBS buffer with the HQ was stopped. Usually, a well-defined, nearly reversible

HQ reduction-oxidation peak pair vs Ag/AgCl (Figure 18, 19) was obtained, due

to the oxidation (+0.35 V) and reduction (-0.15 V) of HQ. A possible increase of

the peak current at negative potentials indicated that oxygen was still present in

the cell, and thus the solution had to be further purged before carrying out the

actual CV measurements. �� 2� � 2��

Figure 18. Typical cyclic voltammetry of 1 mM HQ in pH 7.0 phosphate buffer

using an AuNP/Ab1/Ag/Ab2/MP/HRP (steady: see text) rotating-disk electrode.

Scan rate = 0.2 Vs-1.

53

Figure 19. Typical cyclic voltammetry of 1 mM HQ in pH 7.0 phosphate buffer

using an AuNP/Ab1/Ag/Ab2/MP/HRP eight-electrode array in the microfluidic

channel.

The best experimental conditions allowing us to obtain the lowest

detection limit of the biomarker Nanog were assessed with a single rotating-disk

electrode approach (Figure 20). Afterwards, the method was automated by using

microfluidics.

Rotating-disk electrodes are extensively employed in electroanalytical

research. Advantages of these electrodes over those having different shapes are

the reproducibility of the results, the precise control of the rates at which the

electroactive substances are transported to the electrode (forced convection), and a

practically even current distribution on the surface during the course of the

electrochemical reaction.93

The movement of a liquid caused by a disc rotating around an axis

perpendicular to its plane has been described.94,95 It is assumed that the volume of

the solution in which the disk is present is infinite, the disk is large, and the liquid

flow is laminar. Figure 20 shows the model of movement of the liquid. At large

54

distances from the disk, the liquid moves perpendicula

disk, in the thin layer adhering to the surface of the electrode, the liquid also g

a centrifugal velocity. At very short distances, a layer of quiescent solution is

present, whose thickness does not change in time. In this

effects the electroactive species only by diffusion.

Figure 20. Flow profile that is developed when a circular object is rotated in

solution and how this brings fresh reactant to the surface

Providing the rotation speed is

maintained, the mass transport equation is given by:

where the x dimension is the distance normal to the electrode surface. The mass

transport equation is determined by both diffusion (first term of the right

side member) and forced convection (second term), and both these terms effect

the concentration of the rea

sufficiently negative (positive) to make the red

controlled solely by mass transport, a limiting current (constant in time) arises.

The full and correct equation

rotating disk electrode, was

distances from the disk, the liquid moves perpendicularly to the surface. Near the

disk, in the thin layer adhering to the surface of the electrode, the liquid also g

At very short distances, a layer of quiescent solution is

present, whose thickness does not change in time. In this layer, mass transport

effects the electroactive species only by diffusion.93

Flow profile that is developed when a circular object is rotated in

solution and how this brings fresh reactant to the surface.

Providing the rotation speed is kept within the limits that laminar flow is

transport equation is given by:

dimension is the distance normal to the electrode surface. The mass

transport equation is determined by both diffusion (first term of the right


the concentration of the reagent close to the electrode surface. At a potential

sufficiently negative (positive) to make the reduction (oxidation) process


The full and correct equation, describing the limiting current in the case of

was obtained by Levich96,97:

rly to the surface. Near the

disk, in the thin layer adhering to the surface of the electrode, the liquid also gains

At very short distances, a layer of quiescent solution is

layer, mass transport

Flow profile that is developed when a circular object is rotated in

kept within the limits that laminar flow is

dimension is the distance normal to the electrode surface. The mass

transport equation is determined by both diffusion (first term of the right-hand


gent close to the electrode surface. At a potential

uction (oxidation) process


ent in the case of

55

where i l is the limiting current, n is the number of electron transferred, F is the

Faraday constant, A is the electrode area, D is the diffusion coefficient, ω is the

angular velocity of the rotating disk electrode, c0 is the bulk concentration, and ν

is the kinematic viscosity of the solution.

3.1.3 Analysis of the Immunosensors’ Performance

In our sensor assembly, the working electrode was placed into an

electrochemical cell containing 10 mL of a 1 mM HQ PBS buffer solution, which

was purged with purified Nitrogen for at least 1 h before the measurements. We

used a normal electrochemical cell with 5 holes: one for the working rotating-disk

electrode, one for the reference electrode (SCE), one for the counter-electrode

(Pt), one for the inert gas, and one for the injection of hydrogen peroxide.

Figure 21. In the upper image, there are the electrodes during the incubation of

Ab1. On the right, the rotating disk electrode system is represented entirely, and in

the last image there is a zoom of the electrochemical 5 holes cell that we used for

the measurements.

56

For the sake of better reproducibility, we used 9 electrodes at the same

time: 3 for the control, 3 for one specific concentration of antigen (for example

25 pg/mL) and 3 for another concentration (for example 50 pg/mL). The

electrochemical measurements were carried out using a solution of 1mM of HQ.

The latter is used as the ET mediator to ward hydrogen peroxide reduction, which

is added successfully at 0.4 mM concentration. For the catalytic measurements,

we applied a potential of -0.3 V vs SCE and a rotation rate of 3000 rpm.

Concerning the electrode reaction, it is useful making a few comments

about HRP oxidation by H2O2. The horseradish peroxidase is a protein that

contains a single protoporphyrin IX heme group, and thus the addition of H2O2

converts the iron heme peroxidase enzyme to a ferryloxy species (Figure 21). The

latter can be electrochemically reduced by the electrode.

Figure 22. Oxidation cycle of Horseradish peroxidase.

The immunosensors have a complex sandwich structure. To optimize the

experimental conditions and the value of the detection limit, we changed several

parameters, one at a time, because every biomarker has a specific characteristic

behavior.

At the beginning, we had no information about this protein and thus we

tried almost randomly different concentrations in order to find the best range to

work. This is particularly important also because, beyond a specific concentration,

saturation of the immunosensor occurs and the current does not change with a

further increase of the concentration of the Nanog Antigen (NA). The various

approaches were as described in the following.

57

First approach:

Ab1 25 µg/mL in PBS buffer pH = 7.0 BSA 0.1% in PBS buffer pH = 7.0 NA Several concentrations in PBS buffer pH = 7.0 Biotin-Ab2 1 µg/mL in PBS buffer pH = 7.0 Streptavidin-HRP 5 µg/mL (1:200) in PBS buffer pH = 7.0

As Figure 23 shows, with this method the sensitivity of our immunosensor

is really poor and the signal is not proportional to the concentration of the

biomarker. Thus, we decided to change the conditions by increasing the

concentration of Biotin-Ab2 from 1 µg/mL to 5 µg/mL. We expected an

enhancement of the number of HRP attached to the surface thanks to a larger

number of Ab2 molecules.

Figure 23. Amperometric results of the first approach (for the sake of better

comparison the curves have been arbitrarily shifted).

Second approach:

Ab1 25 µg/mL in PBS buffer pH = 7.0 BSA 0.1% in PBS buffer pH = 7.0 NA Several concentrations in PBS buffer pH = 7.0 Biotin-Ab2 5 µg/mL in PBS buffer pH = 7.0 Streptavidin-HRP 5 µg/mL (1:200) in PBS buffer pH = 7.0

0

50

100

150

0 50 100 150

0pg PBS

25pg PBS

100pg PBS

t, s

58

In this approach, we tried a few concentrations but we did not detect any

particular increase of the signals. Thus, we decided to go back to the

concentration of Ab2 used in the first approach but changed the composition of the

solution for both Biotin-Ab2 and Str-HRP.

Third approach:

Ab1 25 µg/mL in PBS buffer pH = 7.0 BSA 0.1% in PBS buffer pH = 7.0 NA Several concentrations in PBS buffer pH = 7.0 Biotin-Ab2 1 µg/mL in 2% BSA PBS T-20

buffer pH=7.0 Streptavidin-HRP 5 µg/mL (1:200) in 2% BSA PBS T-20

buffer pH=7.0

As Figure 24 shows, the response for the lowest concentrations was good

and the detection limit was particularly low, being 10 pg/mL. On the other hand,

beyond 50 pg/mL, the current did not increase proportionally to the concentration.

We did not know if this was due to the saturation of the sensor or to the presence

of BSA, which acts as a blocking agent not only for the AuNPs platform but also

for Ab1. To address this issue, we repeated this approach while increasing the

concentration of Ab2.

Figure 24. Amperometric results of the third approach (for the sake of better


59

Fourth approach:

Ab1 25 µg/mL in PBS buffer pH=7.0 BSA 0.1% in PBS buffer pH=7.0 NA Several concentrations in PBS buffer pH=7.0 Biotin-Ab2 5 µg/mL in 2% BSA PBS T-20

buffer pH=7.0 Streptavidin-HRP 5 µg/mL (1:200) in 2% BSA PBS T-20

buffer pH=7.0

As Figure 25 shows, the presence of BSA did not allow observing any

difference between the control (0 pg/mL) and the two concentrations employed

(50 and 100 pg/mL). This confirmed that BSA completely blocked Ab1, thereby

prohibiting its interactions with the Nanog biomarker.

Therefore, we decided to change the composition of the solution for both

Biotin-Ab2 and Str-HRP, using PBS T-20.

Figure 25. Amperometric results of the fourth approach (for the sake of better


0

50

100

150

0 50 100 150

0pg BSA T-20

50pg BSA T-20

100pg BSA T-20

t, s

60

Fifth approach:

Ab1 25 µg/mL in PBS buffer pH=7.0 BSA 0.1% in PBS buffer pH=7.0 NA Several concentrations in PBS buffer pH=7.0 Biotin-Ab2 5 µg/mL in PBS T-20 buffer

pH=7.0 Streptavidin-HRP 5 µg/mL (1:200) in PBS T-20 buffer

pH=7.0

As Figure 26 shows, this method seemed to work quite well, but after the

first experiment we could not obtain a similarly low current control; in fact, it was

usually very high, more than the detection limit. This was probably due to HRP

also functionalizing the gold platform, not just Ab2.

Therefore, to better protect the gold platform, we decided to increase the

concentration of BSA during the second step of the procedure, and we prepared a

Str-HRP solution in 0.2 % BSA PBS T-20 buffer pH=7.0.

Figure 26. Amperometric results of the fifth approach (for the sake of better


0

50

100

150

200

0 50 100 150 200

0pg PBS T-20

50pg PBS T-20

100pg PBS T-20

200PBS T-20

t, s

61

Sixth approach:

Ab1 25 µg/mL in PBS buffer pH=7.0 BSA 2% in PBS T-20 buffer

pH=7.0 NA Several concentrations in PBS buffer pH=7.0 Biotin-Ab2 5 µg/mL in PBS buffer pH=7.0 Streptavidin-HRP 5 µg/mL (1:200) in 0.2 % BSA PBS T-20

buffer pH=7.0

As Figure 27 shows, however, in this approach the detection limit was

particularly high, larger than 50 pg/mL.

Since after testing the above approaches we could not establish any

particular method allowing to provide sensitive and reproducible signals, with a

low detection limit, we resorted to modify the preparation of the immunoassay by

introducing magnetic particles. As a matter of fact, Rusling and coworkers nicely

demonstrated that magnetic beads are very helpful to increase the sensitivity of

similar biosensors.9

We also decreased the concentration of BSA in the second step, and we

used 0.1% BSA PBS buffer (pH = 7.0) as the solution for BioAb2-MPs-StrHRP.

Figure 27. Amperometric results of the sixth approach (for the sake of better


0

50

100

150

0 50 100 150

0pg BSA PBS T-20

50pg BSA PBS T-20

200pg BSA PBS T-20

t, s

62

Seventh approach:

Ab1 25 µg/mL in PBS buffer pH=7.0 BSA 0.2% in PBS T-20 buffer

pH=7.0 NA Several concentrations in PBS buffer pH=7.0 BioAb2-MPs-StrHRP 5 µg/mL per 50µL

2.5 µg/mL per 30µL 2500 µg/mL per 20µL

1:10 dilution

in 0.1% BSA PBS buffer pH = 7.0

To amplify the sensitivity and improve the detection limit, we thus

attached multiple HRP labels to carboxylic group on magnetic bead surfaces. HRP

and Ab2 at a 120/1 HRP/Ab2 molar ratio was reached with the carboxylated

magnetic bead using the EDC amidization protocol. Carboxylic acid groups were

first activated using EDC, excess reagent was removed by washing with water,

and the activated particles were then reacted with amine groups on the proteins.

Magnetic beads of diameter ~1 µm provided a very high number of labels on the

surface. After the bioconjugation of HRP and Ab2, the free antibodies and HRP

were easily separated from the Ab2-magnetic-bead-HRP by using a magnet to

localize the beads at one side of an eppendorf and washing to remove unreacted

protein. The amount of active HRP per unit weight of magnetic beads was

determined by reacting the BioAb2-magnetic-bead-StrHRP dispersion with HRP

substrate 2,2’-azino-bis-(3-ethylbenz-thiazoline-6-sulfonic acid) and H2O2. The

total number of magnetic beads, of ~1µm diameter, was 4 x 109 in the dispersion,

provided by Polyscience, and the number of active HRP per magnetic bead was

estimated at 7500.9

As Figure 28 shows, the sensitivity improved very significantly with

respect to the conventional Ab2-HRP AuNPs immunosensor. The current values

of Figure 28 were actually even too high because we wanted to analyze a

significant range of concentrations with no risk of experimental limits. Therefore,

we diluted the initial mixture of BioAb2-MPs-StrHRP.

Figure 29 shows a simple graph in which we compare the slope of the

concentration dependence of the amperometric intensities as obtained with a

single label BioAb2-StrHRP immunosensor with those recorded with the new

immunoassay, based on magnetic particles and BioAb2-MPs-StrHRP.

63



Figure 29. Calibration curves for various approaches, in the two different

immunoarrays.

0

250

500

750

0 50 100 150

0pg Ab2-MP-HRP in PBS 1:10



t, s

0

100

200

300

400

500

0 50 100

Ab2-HRP 2%BSA PBS T-20Ab2-HRP 0.2%BSA PBS T-20Ab2-HRP PBS T-20Ab2-MP-HRP in PBS 1:10Ab2-MP-HRP in PBS 1:20Ab2-MP-HRP in PBS 1:50

Concentration, pg

64

Eighth approach:




1:20 dilution


As Figure 30 shows, the intensities of signals were still high but less than

before, while the slope was equal. Therefore, we diluted the mixture further.



Ninth approach:




1:25 dilution


0

250

500

0 50 100 150




t, s

65

Tenth approach:




1:50 dilution


In the ninth approach, the dilution was not sufficient, and thus we now

make comments directly for the tenth approach in which we observed good values

of current intensity, as Figure 31 shows. The detection limit was 10 pg/mL.

This approach, however, did not display a good reproducibility and thus

we tried to optimize the biological conditions of our immunosensor by employing

solutions with the same pH that was used during the synthesis of Nanog (pH =

7.4). To verify the improvement, we decided to go back to the conventional

biosensor, and we also increased the concentration of BSA in the second step.

Figure 31. Amperometric results of the tenth approach (for the sake of better


0

50

100

150

200

0 50 100 150 200





t, s

66

Eleventh approach:

Ab1 BSA

NA Biotin-Ab2

Streptavidin-HRP

As Figure 32 shows, with this approach we obtained very good results for

sensibility, reproducibility and reliability toward Nanog. In fact, the detection

limit is 25 pg/mL, the sensitivity, calculated as the slope of the calibration plot

(see Figure 33), is 0.639 nA

excellent sensor-to-sensor reproducibility (see Figure 33).

Figure 32. Amperometric results of the eleventh

comparison the curves have been

Eleventh approach:

25 µg/mL in PBS buffer pH=7.02% in PBS T-20 buffer

pH=7.4Several concentrations in PBS buffer pH=7.4

1 µg/mL in 0.1% BSA PBS buffer pH=7.4

5 µg/mL (1:200) in 0.1% BSA PBS buffer pH=7.4



sensitivity, calculated as the slope of the calibration plot

(see Figure 33), is 0.639 nA-mL pg-1, and the small error bars illustrate the

sensor reproducibility (see Figure 33).

Amperometric results of the eleventh approach (for the sake of better


in PBS buffer pH=7.0 20 buffer

4 in PBS buffer pH=7.4

in 0.1% BSA PBS buffer pH=7.4

in 0.1% BSA PBS buffer pH=7.4



sensitivity, calculated as the slope of the calibration plot

, and the small error bars illustrate the

(for the sake of better

Figure 33. Immunosensor calibration plot for Nanog using the eleventh

approach.

On the other hand, we wanted to obtain a detection limit low enough to

have a sensor capable of measuring both normal cancer

with elevated levels of cancer. As we described above, this is possible by using

magnetic beads and thus in the last step we used a 0.1% BSA PBS buffer (pH =

7.4) solution of StrMPs

Twelfth approach:

Ab1 BSA NA StrMPs-bioAb2-bioHRP

As Figure 34 shows, this approach increases the detection limit and the

sensitivity of the sensor very significantly. From the calibration plot (Figure 35)

we obtained a sensitivity equal to 1.98

the conventional immunosensor; the detection limit was 0.1 pg/mL, i.e., ~250

better than the value of 25 pg/mL using the conventional immunosensor.

Immunosensor calibration plot for Nanog using the eleventh


capable of measuring both normal cancer-free patients and patients



7.4) solution of StrMPs-BioAb2-BioHRP.

approach:

25 µg/mL in PBS buffer pH=7.00.1% in PBS buffer pH=7.4

Several concentrations in PBS buffer pH=7.4bioHRP 20 µL

5 µg/mL 50 µL 30 µL of 2.5 mg/mL 10µL is diluted 1:50


(incubated for 30 min)



we obtained a sensitivity equal to 1.98 nA-mL pg-1, i.e., ~3 times larger than using

the conventional immunosensor; the detection limit was 0.1 pg/mL, i.e., ~250


67

Immunosensor calibration plot for Nanog using the eleventh


free patients and patients



in PBS buffer pH=7.0 in PBS buffer pH=7.4 in PBS buffer pH=7.4


(incubated for 30 min)



, i.e., ~3 times larger than using

the conventional immunosensor; the detection limit was 0.1 pg/mL, i.e., ~250-fold


68

Figure 34. Amperometric results of the twelf

comparison the curves have been

Figure 35. Immunosensor calibration plot for Nanog using the twelfth approach.

perometric results of the twelfth approach (for the sake of better


Immunosensor calibration plot for Nanog using the twelfth approach.

(for the sake of better

Immunosensor calibration plot for Nanog using the twelfth approach.

At this point, to better control and automate sample introduction, as well as

steps such as washing, reagent addition, and the amount of H

decided to take the sensor experiment to the microfluidic system.

In microfluidics, the mechanism is similar to that of the rotating

electrode, the main difference being that in this case

stationary while the HQ solution continuously flows into the microfluidic channel

at a determinate speed rate (100

transport regime, H2O

enzyme peroxidase. Figure 36 shows a schematic representation of the

microfluidic system.

Figure 36

The intensity of the a

concentration of NA, and

the flux. Figure 37 shows one representative example of measurement employing

an eight-electrode array. The Nanog concentration was 1 pg/mL. As we can see,

the current does not reach a steady state

method, but increases until reaching a maximum value and then decreases in time

because the concentration of hydrogen peroxide decreases due to the constant flux

of the fluid.


washing, reagent addition, and the amount of H2

decided to take the sensor experiment to the microfluidic system.

In microfluidics, the mechanism is similar to that of the rotating

electrode, the main difference being that in this case the eight-electrode array is


at a determinate speed rate (100 µL/min). By taking advantage of this mass

O2, once injected, can reach the sensor array and


6. Schematic representation of microfluidic system

The intensity of the amperometric current is proportional to the

concentration of NA, and the width of the peak is proportional to the speed rate of


electrode array. The Nanog concentration was 1 pg/mL. As we can see,

the current does not reach a steady state, as with the rotating



69


2O2 injected, we

In microfluidics, the mechanism is similar to that of the rotating-disk

electrode array is


L/min). By taking advantage of this mass

, once injected, can reach the sensor array and oxidize the


Schematic representation of microfluidic system

mperometric current is proportional to the

the width of the peak is proportional to the speed rate of


electrode array. The Nanog concentration was 1 pg/mL. As we can see,

, as with the rotating-disk electrode



70

Figure 37. Amperometric results of 1 pg/mL eight-electrode array using the

twelfth approach.

This was the last result obtained concerning this optimization process.

Further measurements are currently being carried out at the University of

Connecticut. Among them, to obtain a “microfluidic” calibration curve similar to

that of Figure 35 and the use of different NA concentrations in the eight

electrodes.

The results described above demonstrate the utility of gold nanoparticle-

based immunosensors combined with amplified multilabel detection using

magnetic beads for ultrasensitive detection of protein biomarkers. Since a

significant number of biomarkers have normal levels in the low pg/mL range, we

here wished to demonstrate how to improve the detection limit in this range in

order to optimize a sensor capable of measuring both normal (=low)

concentrations representative of cancer-free patients and elevated levels indicative

of cancer. The approach utilizing StrMPs-bioAb2-bioHRP conferred the best

sensitivity and the best detection limit due to the high number of labels on each

magnetic bead. Other advantages of the magnetic beads for multiple labeling

71

include a more narrow and reproducible size distribution than other methods and

ease of preparation of the labeled bioconjugates, featuring magnetic separation of

MPs-Ab2-HRP from unbound proteins by using a magnet.

The increased density of the conductive AuNPs greatly increases the

active surface area for capture antibody attachment to flexible glutathione tethers

and may also be an important factor explaining the improved sensitivity. The key

advantages for higher amperometric signal include densely packed, patternable

conductive nanoparticles resulting in a high surface area, highly conductive

platform with protruding functional groups that allow simple bioconjugation to

large amounts of primary antibodies. The immunosensors required mediation for

the best sensitivity and this is most likely related to the distance between the HRP

labels and the AuNPs, which limits the efficiency of direct ET.

The improvement in detection limit may be also related to better control of

mass transport in the microfluidic system compared to the rotating-disk electrode

approach, leading to better signal-to-noise. The microfluidic biosensor also

showed good stability and reproducibility.9 In general, the advantages of

microfluidics are: i) small volume required [µL, nL] (very important for expensive

reagents and to reduce waste); ii) faster response (reaction time is fast, shorter

diffusion distances); iii) high sensitivity and detection limits; iv) inexpensive

method; v) less energy consumption; vi) capability of rapid multiplexed protein

detection; vi) portable and user friendly; vii) operationally simple; viii) ideal for

point of care diagnostics.

To conclude, this method should be readily adaptable to measure cancer

biomarkers having very low normal levels in serum and tissue, for point-of-care

early detection and monitoring of this disease.

3.2 IRRAS Characterization of Peptide SAM

Infrared reflection absorption spectroscopy (IRRAS) is a useful technique

for studying thin films on reflective substrates. A study of the infrared absorption

spectra of adsorbed molecules may provide detailed information about the

72

structure of the adsorbed species, the nature of the interaction with the surface,

and the strength of the adsorption forces.

3.2.1 The metal-surface selection rule

The IRRAS technique differs from normal IR spectroscopy for a particular

effect given by the reflection of the beam on a surface. Infrared radiation is a

transverse, electromagnetic wave characterized by the orthogonal electric and

magnetic fields oscillating in directions perpendicular to the direction of the wave

propagation. Figure 38 shows a graphical representation of an electromagnetic

wave with the electric field oriented along the y axis. Such a wave, with a well-

defined orientation of the electric and magnetic fields, describes a linearly

polarized radiation and shows that a non-polarized radiation can be regarded as

composed of two orthogonal, linearly-polarized components. The energy of the

magnetic field is equal to the energy of the electric field of an electromagnetic

wave and thus it is convenient to consider only one field, usually the electric one.

When the electromagnetic radiation is incident at a boundary between two phases,

part of the beam is reflected and part of it is refracted into the second medium.

The reflected light combines with the incident light to produce a standing-wave

electric field at the metal surface, where the adsorbing monolayer is present.

Figure 38. Direction of electric and magnetic fields with respect to the

propagation direction of an electromagnetic wave.

73

This interaction between the incident and reflected light causes a phase

change in the electromagnetic radiation that depends upon both the angle of

incidence and the state of polarization of the light. Figure 39 shows the typical

reflection phase shifts expected for a metal. The phase shift for the component

polarized perpendicular to the plane of incidence (s-polarized electromagnetic

radiation) is close to 180° for all angles of incidence, while the phase shift for the

component polarized parallel to the plane of incidence (p-polarized

electromagnetic radiation) changes rapidly at high angles of incidence. The plane

of incidence is defined as the plane that contains the incident and reflected beams.

Figure 39. Phase shift for light reflected from a metal surface. δ|| is the phase

shift for light polarized parallel to the plane of incidence; δ┴ for light polarized

perpendicular to the plane of incidence.

Figure 40 shows the plane of incidence profile of a boundary formed by a

transparent and an absorbing phase. The coordinate system has the x and z axes in

the plane of incidence and the y axis normal to the plane of incidence. The

propagation directions of the electromagnetic radiation are indicated by

dashed lines. The figure shows that the incident beam can be envisaged as

composed of the two linearly polarized orthogonal components: s and p.

-225

-180

-135

-90

-45

00 45 90

Angle of Incidence (Deg.)

Ph

ase

Ch

ang

e o

n R

efle

ctio

n (D

eg.)

74

Figure 40. Reflection and refraction of electromagnetic radiation at the boundary

between a transparent and an absorbing medium. Dashed lines denote the

direction of propagation of radiation. Arrows indicate the electric field vector of

p-polarized light, which lies in the plane of incidence. Symbols and

indicate the advancing and retreating electric field vectors of s-polarized

radiation (which is perpendicular to the plane of incidence). In addition, symbol

shows the (advancing) direction of the y coordinate axis.

Figure 41, instead, underlines the orientation of the electric field vectors of

s- and p-polarized beams with respect to the metal surface for a high angle of

incidence where the phase shift upon reflection is 90°. Due to destructive

interference of the incident and reflected radiation near the metal surface, the

electric field of the s-polarized infrared beam almost vanishes at the metal surface

for all angles of incidence, while the electric field of the p-polarized beam is

enhanced at the metal surface as result of the constructive interference. Note that

at grazing incidence the phase shift is 180° even for the p-polarized component

and, consequently, its incident and reflected electric vectors would cancel at the

metal surface. In other words, a thin layer on a metal reflecting surface should

show some absorption only for the light polarized parallel to the plane of

incidence and for a sufficiently high value of the angle of incidence, but not too

high, otherwise even the p-polarized component undergoes to a destructive

interference.

75

By considering these issues, the IR spectrum for a chemisorbed molecule

may substantially change from that of the same molecule in solution or gas phase.

Because of the reflection, only those modes that have vibrational dipole moments

with components perpendicular to the metal surface will have measurable

intensities. Whereas these normal modes are IR active, the normal modes whose

dipole moment is parallel to the metal surface are IR inactive. This selection rule

characteristic of the reflection adsorption spectroscopy is called “metal-surface

selection rule”. Finally, differences in the IR spectra are also expected for a

substantial change in the degrees of freedom of the chemisorbed species due to

the bond between the molecule and the metal surface. It is well known that an N-

atomic molecule in the gas phase has 3N kinetic degrees of freedom, for a

nonlinear molecule 3 of these are translational, 3 are rotational and (3N-6) are

vibrational modes; for linear molecules there are 3 translational, 2 rotational and

(3N-5) vibrational modes.

p - polarized radiation:

electric field oscillates in the

plane of incidence

phase shift after reflection = 90°

constructive interference

enhancement of electric field at

the metal surface

s - polarized radiation:

electric field oscillates perpendicular to the

plane of incidence

phase shift after reflection = 180°

destructive interference

vanishing of electric field at the metal

surface

Figure 41. Diagrams of reflection of linearly polarized electromagnetic radiation

by a metal mirror for a grazing angle, showing the interaction of s- and p-

polarized beams with the metal surface.98

76

For each non-degenerate vibrational degree of freedom, there is a normal

mode in which all atoms vibrate with the same frequency. However when an N-

atomic molecule is chemisorbed on a solid a new chemical bond is formed

between the molecule and the surface and the adsorbed species is held with a

greater extent of rigidity. This leads simultaneously to a substantial

rearrangements of the bonding pattern between the atoms of the original adsorbed

molecule and, hence, to a changed of the (3N-6) "internal" vibrational

frequencies. In addition, the three translational and the three rotational motions of

the free molecules are usually converted to vibrational modes arising from the

presence of the new chemical bonds. These new modes may become substantially

mixed with the modified internal vibrations of the adsorbed molecule. Therefore,

it is generally strictly meaningful to speak of a total of 3N new modes associated

with the system of adsorbate plus adsorbent. In some cases, the torsional mode of

the chemisorbed species may be approximate to a free rotation of the molecule

around the new bond formed; then, there will be (3N-l) new vibrational modes. In

addition to changes in the degree of freedom of the molecule, new modes of

vibration can arise from the coupling between the vibrations of the chemisorbed

species and the lattice modes of the surface. This particular coupling can add other

3N vibrational degrees of freedom.99

3.2.2 Characterization of Aib-Homopeptides

Before addressing the actual IRRAS measurements carried out on the

peptide SAMs prepared on well activated and characterized gold surfaces (see

Experimental), it is important to assess and comment the IR spectroscopy features

of the free peptides in solution.

The FT-IR absorption spectra of the thiolated Aib-peptides, S-protected by

the trityl (Trt) group, were obtained in CH2Cl2, which is a solvent of low polarity

having no propensity to behave as a hydrogen-bond acceptor. The analysis was

focused on the N-H and amide stretch modes.

The N-H stretch region (Amide A region, ranging from 3500 to 3200 cm-1)

is composed by a high-energy region corresponding to the NH groups that are not

involved in hydrogen bonds and a low-energy region pertaining to NH groups

77

involved in intra- and inter-molecular hydrogen bonds (Figure 42 and 43). The

amide region ranges from 1700 to 1500 cm-1 and contains two main signals: the

one higher in energy is called Amide I and is principally related to the stretch of

the C=O groups. The other, lower in energy, is called Amide II and is more

complex with an important contribution from N-H bending and C-N stretching

vibrations (Figures 44 and 45).100-101

As we can see in Figures 40 and 41, for both the plus and minus series

there is a red-shift of the Amide I signal, as the peptide chain is made longer, as

previously reported for similar peptide series by Toniolo et al.100

In particular, these authors found that this red-shift is optimized with the

octamer because of formation of a fully developed, stable intramolecular H-bond

network. Interestingly, we can observe a shoulder in the amide I of 1- at high

energy. It thus appears that while for the 1+ the H-bond is more stable and the

signal is prevalently related to the H-bonded C=O stretch, in the minus series the

H-bond is weaker and the C=O component not involved in such interaction is

more evident.

Figure 42. The amide A region for the plus series.

78

Figure 43. The amide A region for the minus series.

Figure 44. Amide I and amide II regions for the plus series.

b

)

79

Figure 45. Amide I and amide II regions for the minus series.

Concerning the amide A region (Figure 42 and 43), there are two signals: a

less intense signal at higher frequency and a more intense one between 3350 and

3300 cm-1. The first is related to free N-H groups. According to the 310-helix

features, two N-H and two C=O groups are not involved in intramolecular H-

bonds, as highlighted in Figure 46. As reported by Toniolo et al.100, the intensity

of the band related to the stretch of the N-H bonds involved in intramolecular H-

bonds increases as the peptide is made longer. The red-shift of the band tends to

become independent of the peptide length for sufficiently long peptides, meaning

that the 310-helix is fully formed and stable. With our systems, we observed the

same intensity increase and red shift. The intensity of the band related to the free

N-H groups does not change because the number of those groups (two) does not

vary with the peptide length.

Figure 46. The C=O and the N-H not involved in the intramolecular H-bonds

(plus series) are circled in blue and orange, respectively.

80

To verify the extent of stability of the 310-helices formed by the peptides of

the plus and minus series, we applied the same analysis previously described for

these systems. The ratio between the integrated intensity of the band of H-bonded

N-H groups to free N-H groups, divided by the number of actual H-binds, should

display a dependence of on the number of intramolecular H-bonds that tends to

level off for sufficiently long peptides: this is an indication of formation of stable

and stiff 310-helices.100

Figure 47 compares the results obtained with the plus and minus series in

comparison with the data previously reported for Z-(Aib)n-OtBu (Z =

(benzyloxy)carbonyl, OtBu = tert-butoxy, n = 3-11).100 If the values at large H-

bond values are taken as a reference, one can estimate how the percentage of

intramolecular H-bonds changes as the peptide is made longer. This is illustrated

in Figure 48. The shorter oligomers, as expected, display less stability but for

longer peptides of the investigated series we can conclude that the helices are

essentially fully formed, which implies stiffer peptides.

Figure 47. Dependence of the normalized ratio between the integrated intensity of

free and intramolecularly bonded N-H groups on the number of H-bonds (for

explanations, see text).

H-Bonds

0 2 4 6 8 10

[Are

a N

Hbo

nd/A

rea

NH

free

]/nH

1

2

3

(Areaintra/Areafree)/nH Toniolo (Areaintra/Areafree)/nH Trt plus series(Areaintra/Areafree)/nH Trt minus series

81

Figure 48. Dependence of the peptide stiffness on the number of H-bonds.

We should finally note, however, that the FT-IR technique alone does not

allow one to unequivocally establish the specific helical structure adopted by the

peptides. Full assessment also requires extensive characterization by NMR

spectroscopy. For the peptides here investigated, this aspect was previously dealt

with, and will not be discussed further.83 Here, if suffices to stress that the

presence of the 310-helical structure was fully confirmed for most of the peptides

here investigated.

3.2.3 Characterization of Aib-Homopeptides SAMs

Vibrational spectroscopy is a powerful technique for studying peptide

structures and interactions. In the previous section, we discussed the IR absorption

behavior of the free (protected) thiolated peptides and, particularly, how the N-H

stretch region (amide A) allows us to distinguish between the N-H groups that are

involved in intramolecular hydrogen bonds and those not involved in such

interactions. We also saw that other important information on the 310-helix

H-Bonds

0 2 4 6 8 10

% H

-Bon

ds

50

60

70

80

90

100

110

plus series (Trt)minus series (Trt)Toniolo paper values

82

structure and stability can be obtained from the amide I (and amide II) region.

IRRAS spectroscopy was utilized to obtain information on the structure of the

peptide SAMs and to further investigate their stability. In fact, the same

information provided from FTIR of free ligands can be obtained from the position

and intensity of the IRRAS peaks. In addition, information on the organization of

the monolayer can be gathered by taking advantage of the metal-surface selection

rule.

The two series of peptides, self-assembled onto carefully annealed gold

surfaces (see Experimental) were thus studied by IRRAS; the spectrum of 0+,

however, could not be satisfactorily analyzed because of its week signals. Figure

48 shows the full spectrum of the SAMs formed from the peptides of the plus

series. The presence of bands that clearly increase as the peptide length increases

is evident. We will first comment the Amide A region, and then the Amide I and

II regions.

Plus Series:

Figure 49. The FTIR-RAS spectra for the plus series correspond (bottom to top)

to 1+ (pink), 2+ (green), 3+ (blue), 4+ (red), and 5+ (black).

Wavenumber (cm-1)

15002000250030003500

Abs

orba

nce

-8e-4

-6e-4

-4e-4

-2e-4

0

2e-4

4e-4

6e-4

8e-4

1e-3FTIR 5+ FTIR 4+ FTIR 3+ FTIR 2+ FTIR 1+

83

Hydrogen bonding and the coupling between transition dipoles are among

the most important factors governing conformational sensitivity of the amide

bands.100

The N-H stretch region, from 3200 cm-1 to 3500 cm-1, clearly shows

(Figure 50), that compared to the free molecules the band corresponding to the

two free N-H groups disappears, which means that all the N-H groups are bonded

either intra- or inter-molecularly. In fact, we should consider that while in the

SAMs the peptides are forced to organize close to each other. In this way, the free

N-H are conceivably involved in intermolecular interactions and thus the high-

frequency band disappears. This conclusion also implies that neighbor peptide

adsorbates should provide C=O groups that are already involved in intrachain

C=O···H-N hydrogen bonds. These three-center hydrogen bonds allow formation

of a strong interchain bonding network and, since for the plus series they are

embedded inside of the monolayer, they are largely unaffected by an increase of

the peptide length, as already observed for the corresponding 3D SAMs formed

on gold nanoclusters.83

The Amide A region also shows that as the peptide chain length increases

the frequency of the N-H stretch band of intramolecularly H-bonded peptides

decreases, which is a marker of the increase of peptide stiffness (see Figure 51).

As we discussed in the previous section, based on the work of Toniolo and

coworkers one can relate the percentage of intramolecular H-bonds of the 310-

helix type to the number of residues.100

We applied the same treatment to the trityl-protected thiolated peptides of

the plus series and could estimate that even with 1+ helicity is more than 70%,

eventually becoming 100% for the longest peptide. These data highlight the

outstanding efficiency of the Aib unit in stabilizing folded and helical species

even in very short peptides, but also explain the broadness of the IR bands, which

result from a combination of contributions from multiple conformations.83

Since the helicity percentage requires to compute a factor including the

integrated absorbance of the free N-H groups, this calculation cannot be applied to

the peptides in the SAM. On the other hand, the very fact that the peptide is inside

the monolayer implies less freedom degrees and thus the helix is expected to be

even stiffer than in solution. This is also known for simple alkanethiols, where the

84

all-trans conformation is stabilized by Van der Waals interactions with neighbor

molecules in the SAM.56 We are thus confident that not only the 310-helix forms,

as already inferred from the IR spectra of the free molecules in solution, but also

that it is probably even stiffer.

From the comparison between the amide A band of the free peptides and

those self-assembled on the gold surface, another interesting aspect emerges (see

Figure 51). In general, we notice that the peptide-length dependence of the band

maximum for the free (trityl-protected) peptides in solution and that for the self-

assembled peptides is approximately similar. Figure 48 shows that the

wavenumber of the amide A band for the plus series linearly decreases with an

increase of the number of H-bonds. The main difference between the two series is

an almost constant shift: the plus series of chemisorbed peptides is shifted toward

lower wavenumbers by ~4 cm-1. This is probably related to what said before:

while in the SAM, the peptides are stiffer.

Figure 50. Amide A band for the peptide SAM of the plus series, with the only

exception of 1+ because it had a different behavior, so it was treated later.

Wavenumber (cm-1)

3200330034003500

Abs

orba

nce

0

1e-4

2e-4

3e-4 2+ 3+4+ 5+

85

Figure 51. Comparison between the wavenumbers of the amide A band for the

plus series of the free protected peptides (blue color) and for the plus series of the

self assembled peptides (red color).

The band of 2+ shows the presence of two components. The low-

frequency component appears to be present also when the peptide is made longer

(the overall bad is very broad), but the red shift of the main band prevents its

detection for 4+ and 5+. The low-frequency component is typical of

intermolecular H-bonds, as already found and discussed in detail with Aib-peptide

protected gold nanoclusters.83 We here made the assumption that the low-

frequency component of 2+ is present for all peptides.

To better appreciate the relative similarities and differences, also in terms

of absorbance maxima, we thus normalized the N-H stretch region (Figure 53) by

taking into account the number of molecules per cm2 (Table 4). These data were

calculated using the average of the charge (µC) per cm2 obtained in reductive

desorption measurements.89 By this approach, one can take into account the actual

difference between the surface concentrations. In other words, the low-frequency

band of 2+ could be normalized for the concentration and this value was

subtracted from the normalized broad band of the other peptides. The contribution

H-Bonds

0 1 2 3 4 5

Wav

enum

ber

(cm

-1)

3320

3340

3360

3380

86

of the N-H band of intermolecularly H-bonded peptides was obtained by

deconvolution89b of the spectrum of 2+, as illustrated in Figure 52.

Table 4: Values of Surface Coverage.

H-Bonds Plus Series (mol cm-2) Minus Series (mol cm-2)

0 4.15 ± 0.25 · 10-10 -

1 4.03 ± 0.24 · 10-10 6.40 ± 0.70 · 10-10

2 3.84 ± 0.20 · 10-10 -

3 4.99 ± 0.43 · 10-10 3.70 ± 0.05 · 10-10

4 4.59 ± 0.34 · 10-10 -

5 4.31 ± 0.26 · 10-10 3.14 ± 0.12 · 10-10

This deconvolution clearly shows that the signal can be deconvoluted as

the sum of two non-linear least square Gaussian fits, one due to the intramolecular

bonds at 3349 cm-1 and the other due to the intermolecular bonds at 3308 cm-1.

The latter frequency is as found with Aib-peptides protecting gold nanoclusters89:

in the end, we are comparing very similar systems, the main difference being that

small Au clusters tend to be roughly spherical (3D SAM).

Figure 52. Deconvolution of amide A region of 2+.

Wavenumber (cm-1)

3200330034003500

Abs

orba

nce

0

2e-5

4e-5

6e-5

87

Figure 53. FTIR-RAS spectra for the amide A normalized by taking into account

the concentration and the presence of the N-H band of intermolecularly H-bonded

peptides.

As already noted, shorter oligomers exhibit in solution a small percentage

of intramolecular H-bonds. In SAMs, on the other hand, the presence of neighbor

molecules reduces the freedom degrees of each molecule and should thus make

the helix more stable and stiffer. This is true for most Aib peptides. Notably,

however, Figure 51 shows that the amide A band of 1+ does not have the classical

signal related to an intramolecular H-bond (according to Figure 48, a band at

~3360 cm-1 was predicted). Although weak, the only detectable bands are at lower

energies. These frequencies are typical of intermolecular H-bonds, as already

discussed. It appears, therefore, that when 1+ is self-assembled on a gold surface

it undergoes a significant change of its secondary structure. The free energy of the

system would thus be lowered when the intramolecular (310-helix like) H-bond is

broken in favor of intermolecular H-bond(s).

Wavenumber (cm-1)

3200330034003500

Abs

orba

nce

0

1e-4

2e-4

3e-42+ intra3+ intra4+ intra5+ intra

88

Figure 54. Deconvolution of amide A region of 1+.

As already noted, information on the peptide conformation can be obtained

from the amide I and II regions (Figure 55). We remind that the fine structure of

these bands is quite complex because they are not pure modes but, instead, consist

of several components. Let us focus our attention prevalently on the comparison

of the two amide bands, the main reason being that from the ratio between the

corresponding intensities the tilt of the peptide adsorbates can be estimated.

Whereas the intensity of the amide I band, which corresponds mostly to

the C=O stretch and occurs at 1680 – 1672 cm-1, increases with the number of

residues for the plus series, the intensity of the amide II band, centered at ~1530

cm-1, is almost constant (see Figure 55). The main frequency is roughly constant

for peptides 3+, 4+ and 5+, but the value decreases for peptides 1+ and 2+ (see

Figure 56). The different sensitivity of the two bands on the peptide length are

related to the metal-surface selection rules, according to which the transition

moment related to the amide I band is more perpendicular to the surface, while the

transition moment related to the amide II is only mildly affected.

Wavenumber (cm-1)

3200325033003350

Abs

orba

nce

-1e-5

0

1e-5

2e-5

3e-5

89

Figure 55. IRRAS spectra of the amide I and II bands for the plus series.

Concerning the Amide I band of the free and the SAM peptides (Figure

53), the frequency of the signal slightly decreases as the number of hydrogen

bonds increases. This behavior is related to a progressive stabilization of 310-helix

with an increase in the number of residues.100 Unlike amide A, the amide I band

of the self-assembled peptides is shifted by ~4 cm-1 to higher wavenumbers than

in the corresponding free peptides. This behavior is quite intriguing because it

would point to a less strong helical structure of the peptides while in the SAM.

This is in contrast with the amide A behavior, the latter being easily

understandable because of the reduced freedom degrees inside the SAM. At the

present stage of investigation, we have no explanation of the different trend

showed by the amide I data. As for the Amide A region, 1+ displays some

different features from the other peptides of the plus series. This is evident from

the shape of the amide I band (Figure 57) that is composed by more than one

component. We will see that a similar complex situation is also displayed by 3-

and, particularly, 1-. The complex pattern of the Amide I band confirms our initial

hypothesis that when 1+ is self-assembled on a gold surface it undergoes a

significant change of its secondary structure.

Wavenumber (cm-1)

1500160017001800

Abs

orba

nce

-0.0002

0.0000

0.0002

0.0004

0.0006

0.0008

0.00105+ 4+ 3+ 2+ 1+

90

Figure 55. Comparison between the wavenumbers of the amide II band for the

plus series of the free protected peptides (blue) and for the plus series of the self-

assembled peptides (red).

Figure 56. Comparison between the wavenumbers of the amide I band for the

plus series of the free protected peptides (blue color) and for the plus series of the

self assembled peptides (red color).

H-Bonds

0 1 2 3 4 5

Wav

enum

ber

(cm

-1)

1500

1510

1520

1530

1540

H-Bonds

0 1 2 3 4 5

Wav

enum

ber

(cm

-1)

1665

1670

1675

1680

91

Figure 57. Deconvolution of amide I region of 1+.

The frequency of the main bands of the plus series, for both the Amide A

and the amide carbonyl regions, are gathered in Table 5.

Table 5: Values of wavenumbers obtained for each peptide for the plus series.

H-Bonds Region Wavenumber (cm-1)1

1 Amide I 1679.6 (1668.1, 1690.6) Amide II 1511.6 Amide A 3305.2 (3219, 3277)

2 Amide I 1679.0 Amide II 1524.7 Amide A 3349.8 (3308.2)

3 Amide I 1677.1 Amide II 1530.4 Amide A 3334.7



1 The frequency values obtained from deconvolution are in parenthesis.

Wavenumber (cm-1)

1600165017001750

Abs

orba

nce

-5.0e-5

0.0

5.0e-5

1.0e-4

1.5e-4

2.0e-4

2.5e-4

92

Minus Series:

Figure 58. The FTIR-RAS spectra for the minus series correspond (bottom to top)

to 1- (blue), 3- (red), and 5- (black).

We now focus on the minus series. This series consistently displays

absorbances significantly lower than those of the corresponding plus series. This

is related to both a lower surface concentration (see Table 5) and other factors, as

explained in the following.

The spectral region pertaining to the amide A band (Figure 59) shows that

1- has a high-energy band in the region of free N-H bonds, and both 1- and 3- do

not show the intramolecular band at the expected wavenumbers, but show a broad

band at lower energies that most likely corresponds to the stretch of

intermolecularly-bonded NH groups. We should mention, however, that for 3- this

behavior is less reproducible, as if for this peptide the SAM structure is very much

dependent on the preparation of the SAM. We believe (see below) that this is

related to this peptide being near the “transition” between forming 310-helices and

other, more intramolecularly H-bonded structures. For 5-, however, we observe a

Wavenumber (cm-1)

15002000250030003500

Abs

orba

nce

-2e-4

-1e-4

0

1e-4

2e-4

3e-4

4e-4

5e-4FTIR 5- FTIR 3- FTIR 1-

93

behavior very similar to that of peptides from 2+ to 5+, albeit with a quite lower

intensity.

Whereas the spectral region pertaining to the amide I and II bands (Figure

60) does not show significant differences for 3- and 5- with respect to the plus

series (beside the intensity of the signals), 1- displays a band that is quite different

from that of the other peptides.

It appears, therefore, that peptides 1+, 3- and, particularly, 1- undergo a

heavy conformational change when organized in SAMs on a gold surface. This

was already emphasized in previous work from this laboratory, showing that the

packing degree of 1- is particularly large.89 The value, 6.40 ± 0.70 · 10-10 (Table

4), can indeed be compared with the surface coverage typical of alkanethiols’

SAMS on Au (111) lattice, i.e., 7.76·10-10 mol cm-2.102

Generally, for the other peptides the packing is quite smaller because the

peptides have a more complex and bulky structure than a simple alkanethiol

chain. The coverage calculated for 1-, however, is as much as 82% the alkanethiol

value, witnessing a very similar packing and suggesting that for this particular

peptide the helical structure is probably lost in favor of a more elongated structure

(note: ideal alkanethiols’ SAMs are formed by molecules adopting an all-trans

conformation).

The frequency of the main bands of the minus series, for both the Amide

A and the amide carbonyl regions, are gathered in Table 6.

Table 6: Values of wavenumbers obtained for each peptide for the minus series.

H-Bonds Region Wavenumber (cm-1)1

1 Amide I 1685.8 (1669.8; 1699.19) Amide II 1556.5 Amide A 3421.3



1 The frequency values obtained from deconvolution are in parenthesis.

94

Figure 59. Amide A band of the peptide SAM of the minus series.

Figure 60. Amide I and II band of the peptide SAM of the minus series.

Wavenumber (cm-1)

31003200330034003500

Abs

orba

nce

0

1e-4

2e-4

3e-4

Wavenumber (cm-1)

1500160017001800

Abs

orba

nce

-0.0002

0.0000

0.0002

0.0004

0.0006

0.0008

0.00105-3-1-

95

Based on the IRRAS observations, we can thus conclude that while in the

SAM the plus series affords a more stable secondary structure than the minus

series. The shortest peptides give less stable 310-helices, probably preferring to

form intermolecular H-bonds. Disruption of the 310-helical structure is more

evident with the minus series.

3.2.4 Determination of the Orientation of the Absorbed

Molecules

The orientation of the absorbed molecules within the SAM is determined

by several factors such as the adsorbate−substrate binding strength, the geometry

of the binding site, the intermolecular interactions between the adsorbed

molecules, and the solvent−adsorbate interactions.103

As a result, the orientation of the peptide helix axis with respect to the

substrate's surface normal is better described as a distribution of values. An

orientation distribution function is thus necessary to interpret the IRRAS spectrum

and calculate the average tilt angle from band intensities. Thus, Samulsky and

coworkers104 employed a Gaussian orientation distribution function to infer the

peptide helix axis orientation in the SAM on gold. The helix axis orientation was

obtained by optimizing the agreement between theoretical values for the amide I

to amide II absorbance ratio, D (see equation 7), and the observed D value from

the IRRAS experiments.

Figure 61 shows the coordinate system that describes the interaction

between the polarized, incident IR electric field and the molecular transition

moment in the grazing angle IRRAS experiment. Due to the surface selection

rule105, the only active electric-field component of the incident light (n) is parallel

to the Z-axis of the laboratory system (X, Y, Z), the gold surface being in the

X−Y plane. The transition moment m (amide I or amide II band) is located in a

molecule (helix)-fixed frame (x, y, z) by its respective polar and azimuthal angles

(α, β). The x, y, z frame is oriented with the z-axis along the helix, and that axis is

located in the laboratory system by θ, and Φ. γ is the azimuthal angle about the

96

helix axis. Using the relations among the Euler a

is given by:

Figure 61. Coordinates of the laboratory (X, Y, Z) and molecular (x,

for a tilted peptide on gold surface.

The IRRAS absorption intensity,

uniform azimuthal distribution about the helix axis

over the interval [0, 2π]

The α-dependence is simultaneously averaged by this assumption for

the angles between the transition moment and the helix axis were determined to

be 39° for β1 (amide I band) and 75° for

A(θ), where θ describes the tilt angle between the helix axis and the surface

normal.

helix axis. Using the relations among the Euler angles106, the projection of

Coordinates of the laboratory (X, Y, Z) and molecular (x, y, z) systems

peptide on gold surface.

RAS absorption intensity, A, is proportional to |m·n|2. Assuming

uniform azimuthal distribution about the helix axisγ is uniformly distributed

Samulski found that

dependence is simultaneously averaged by this assumption for


(amide I band) and 75° for β2 (amide II band),107 A can be written as

describes the tilt angle between the helix axis and the surface

the projection of m on n

y, z) systems

. Assuming

is uniformly distributed

dependence is simultaneously averaged by this assumption for γ. As


A can be written as

describes the tilt angle between the helix axis and the surface

97

The observed IRRAS absorption intensity (Aobs) depends on the orientation

distribution of the helices in the film. Thus, Aobs is the weighted superposition of

contributions, A(θ) W(θ) dθ, where W(θ) is the normalized helix axis orientation

distribution function. Hence, we have

�5� �� |� · �|�� !

�!

Similarly, the average of any function X of the tilt angle of the helix axis

relative to the surface normal is

�6� �#� � � #�� !

The amide I to amide II absorbance ratio is

�7� % � & ��'��'' � & ( �|� · �|��' �� !( �|� · �|��''�� !

where K is the scaling factor that relates the intrinsic “oscillator strength” of the

amide I and amide II vibrational modes. K was determined to be 1.35 ± 2 from the

transmission FTIR spectrum of a peptide KBr pellet.104

The Gaussian orientation distribution function that was used to model the

helix axis orientation distribution is

�8� �� *�+�,- ./ 121� �� / �!��2 sin��

where N is the normalization constant:

�9� * � � �,- �! ./ 121� �� / �!��2 sin��

98

Assuming the monolayer to be an ideal isotropic film with a uniplanar

distribution of helix axis confined to the gold substrate, θ results constant.

Therefore, the helix-axis orientational distribution is a delta function: W(θ) = δ(θ -

90°). This distribution is certainly reasonable for our monolayer.

The observed ratio of the amide I and amide II integrated intensities, Dobs =

A Iobs/A

IIobs, is thus calculated from the equation, for which we know W(θ) = δ(θ -

90°). Therefore, the orientation of the peptides adsorbed on the gold surface was

determined according to the following equation, assuming an isotropic monolayer:

�10� % � & ��'��'' � & �|89 · :|��|899 · :|�� 1.35 �3 cos� � / 1��3 cos� ?+ / 1� � 2�3 cos� � / 1��3 cos� ?� / 1� � 2

�11� @% � A B%B�'A · @�' � A B%B�''A · @�'' � 1�'' · @�' � �'��''�� · @�''

A Iobs and AII

obs are the observed absorbances of amide I and amide II

bands, θ is the tilt angle of the helical axis, and β1 and β2 represent the angles

between the transition moment and the helix axis, those value were taken to be

39° and 75°, respectively.107

In order to take the maximum value of observed absorbance, we selected

the two zones of the spectrum where the amide I and amide II bands are present.

Then, we employed three different nonlinear regression analyses (the Gaussian,

the Pseudo-Voigt, and the Lorentzian fitting methods) to optimize the agreement

between the theoretical values for the amide I to amide II absorbance ratio, and

the observed D value from the IRRAS experiments. After several tests, we opted

for using only the Gaussian regression, which gave the best peaks. In some cases,

to find the best values of maximum absorbance, we also used a deconvolution,

and to calculate the corresponding error, we used the conventional linear

propagation:

�11� C � C! � D · �.�!.EFG�GH� IJ2

99

�12� @C � @C! � �.�!.EFG�GH� IJ2 · @D � KD�, / ,!��.�!.EFG�GH� IJ2LM N · @L� KD�, / ,!��.�!.EFG�GH� IJ2LM� N · @,!

Then, to calculate the tilt angle of each peptide and the corresponding

errors, we used the following equations, obtained from equation 10.

�13� � � DOPQR STU 2& / 2%% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�& � 1V · 13W

�14� @� � YZ[/ 1\1 / U 2& / 2%% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�& � 1V · 13

· 12 1\U 2& / 2%% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�& � 1V · 13· 13 ] 2% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�&/ �3 cos� ?� / 1��2& / 2%��% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�&��^_ · @%

� YZ[/ 1\1 / U 2& / 2%% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�& � 1V · 13

· 12 1\U 2& / 2%% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�& � 1V · 13· 13 U 2% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�&� &�2& / 2%��% · �3 cos� ?� / 1� / �3 cos� ?+ / 1�&��V_ · @&

100

From these calculations, we obtained the values shown in Table 7.

Table 7: Amide I to Amide II Absorbance Ratio and relative Tilt angle.

H-Bonds D = AI/AII Tilt Angle θθθθ (degrees)

1+ 1.16 54.7 2+ 1.68 49.5 3+ 2.06 44.6 4+ 2.80 37.9 5+ 5.90 22.4

1- 1.20 57.8 3- 1.25 56.7 5- 1.38 54.2

Of course, all these data rely on a series of assumptions. First, we assume

of having a monatomically smooth Au(111) surface and that there is no surface

roughness (i.e., an angular spread resulting from a less-than-flat surface that

should be convolved with the angular distribution of the helix axes). This is a

significant assumption as the IR beam size is macroscopic in areal dimensions and

it is doubtful that a uniform surface (other than a single crystal) describes such a

large area. On the other hand, in the Experimental section we stressed that the

freshly annealed surface is mostly a (111) surface and the roughness factor is

close to 1, being 1.16 ± 0.08. Second, we assume that the SAM is isotropic and

defect-free, and this is something that is obviously not true as thoroughly

commented upon in the introducing chapter. Finally, how well is the helix axis

defined? At the polypeptide-air interface, for example, we expect some dynamic

behaviour that introduces some uncertainty about the definition of the helix axis

itself. That the helices are stable is true for the longest peptides of the plus and

minus series, but it is not completely valid for shorter peptides, particularly those

of the minus series. For these peptides, 1θ and 2θ may thus be slightly different.

Because of these assumptions, we consider that an error of at least ±5° is more

reasonable than that of ±1° deriving from our calculations. Figure 63 shows that

the long peptides of the plus series are less tilted than shorter ones.

We should stress that the figure also includes 1+ and 1-, although the error

on the determination of their amide II intensity indicates that the calculated tilt

101

angles suffer an even larger error than that indicated in the figure. Noteworthy,

Figure 63 shows that when the number of H-bonds increases the tilt angle

decreases smoothly. Therefore, it appears that long peptides are suitable to better

self-assemble on the surface. In other words, the more rigid is the secondary

structure of the peptide, the more the SAM is ordered and the molecules are

perpendicular to the surface. It is also interesting to note that in the minus series

the tilt angle depends very little on the peptide length. These angles are similar to

those of 1+ and 2+. This means that the peptides of the minus series are, on

average, significantly more tilted than those of the plus series.

The tilt angles that we calculated are generally larger than those of

alkanethiols, ~30°. On the other hand, helical peptides show tilt angles in the

range from 20° to 55°,108,109,110,111 in line with our results. The majority of

literature calculations are based on IRRAS data.

Figure 62. Dependence of the ratio between the absorbance of amide I and that of

amide II on the number of H-bonds for the plus and the minus series.

H-Bonds

0 1 2 3 4 5

Rat

io A

mid

e I/I

I

0

1

2

3

4

5

6

7Plus Series Minus Series

102

Figure 63. Dependence of the tilt angle of the helix axis on number of H-bonds

for the plus and minus series.

3.2.5. Determination of the Thickness of the Monolayer

The thickness of the monolayers (h) can be calculated from the tilt angle

(θ) and the length of the peptide (l). The lengths of the investigated peptides have

been calculated (for perfect 310-helices) from molecular models. The length was

calculated from the hydrogen of the SH group (mimicking Au) to the hydrogen of

the most distant CH3 of the tert-butyl group. The calculated lengths are: 1+, 11.6

Å; 2+, 13.6 Å; 3+, 15.2 Å; 4+, 16.7 Å; 5+, 18.6 Å; 1-, 11.3 Å; 3-, 15.2 Å; 5-, 18.9

Å. As shown in Figure 64, the thickness of the monolayer can be calculated by

applying a simple trigonometric formula. This calculation logically brings with it

all the approximations made with the calculation of the tilt angle. Although the

calculated values are affected by a non-negligible error, a relative comparison

between peptides of the plus and minus series can still be carried out.

H-Bonds

0 1 2 3 4 5

Tilt

Ang

le

10

20

30

40

50

60

70

Plus SeriesMinus Series

103

Further investigations on the monolayer thickness have been planned for

future work using other methods of analysis, such as ellipsometry and ARXPS

(angle resolved X-ray photoelectron spectroscopy: this is a technique in which the

photoelectron take-off angle allows reducing the depth from which the XPS

information is obtained; its most important application is, in fact, the estimation of

the thickness of thin films and thus SAMs), and the values will be compared. This

comparison will provide a useful way to validate the method based on the amide I

and amide II ratio.

As Figure 62 shows, the thickness of the monolayers for the plus series

increases with the number of H-bonds but, due to the decrease of the tilt, not

according to a linear function. For the minus series it appears that the thickness

increases in a similar way, but with an average smaller slope. To conclude, the

minus series forms thinner monolayers than the plus series. If we now consider the

surface coverages of Table 4, a smaller thickness of these monolayers is probably

determined by a less effective packing on the surface, which may allow the

peptides to tilt more significantly.

l

αcos⋅= lh

104

Figure 64. The graph shows the trend of the monolayer thickness as a function of

the number of hydrogen bonds for the plus (blue circles) and the minus series (red

circles). The inset shows a graphical representation showing how the calculation

was carried out.

105

4. Conclusions

Nowadays the Enzyme-Linked Immunosorbent Assay (ELISA) based on

optical absorbance is the standard bioanalysis method with claimed detection

limits as low as 3 pg/mL for Prostate Specific Antigen (PSA).112 ELISA, however,

suffers limitations in analysis time, sample size, and multiplexing. Preliminary

studies in Jim Rusling’s laboratories suggest good potential for nanoparticles to

make bioelectronic sensor array platforms to measure collections of cancer

biomarkers simultaneously, at high sensitivity (for example the detection limit for

PSA was obtained as low as 0.5 pg/mL)9, and without compromising analysis

time or sample size. To assess the accuracy of the AuNP sensor in real biomedical

situations, a collection of human serum samples with varying PSA content were

used. These samples were also assayed by standard ELISA method. As Figure 65

shows, immunosensor results indeed display a very good correlation with ELISA

from sub-ng/mL concentration range to above 10 ng/mL, for this representative

serum samples set.9

Figure 65. Validation of AuNP sensor results for human serum samples by

comparing against results from ELISA determination for same samples.

106

It was thus demonstrated that the AuNP platform is competitive with other

state-of-the-art approaches in terms of both sensitivity and detection limit in real

samples, and may prove to have marked economic advantages in future array

fabrication. On these grounds, we employed this kind of electrochemical

immunosensor to a new biomarker that may be involved in carcinogenesis of

cervix and progression of cervical carcinoma. We should stress, in fact, that

nowadays researchers still do not know the detection limit of this particular

biomarker; similarly, the difference of concentration between healthy individuals

and patients with cancer is still unknown.

Since we did not know the range of the expression of the Nanog

biomarker, we started by adopting the same approach that was previously adopted

for the PSA biomarker.9 Step by step, we optimized the experimental conditions

by changing the concentrations of the several solutions used for preparing the

immunosensor. Inhibition of non-specific binding of labeled detection antibodies

was the most crucial step, because if the concentration of BSA was too high the

signal was annihilated, but if the concentration was too low the signal was not

reproducible.

Since a significant number of biomarkers have normal levels in the low

pg/mL range, we aimed at reaching sensitivity in this concentration range in order

to make a sensor capable of measuring both normal (= low) concentrations

representative of cancer-free patients and elevated levels indicative of cancer.

Therefore, we implemented our immunosensor using StrMPs-bioAb2-bioHRP that

conferred the best sensitivity and the best detection limit, due to the high number

of labels on each magnetic bead. In fact, as demonstrated in the Results and

Discussion, we obtained a calibration plot that shows high sensitivity, good

reproducibility and the low detection limit of 0.1 pg/mL; this will allow the study

of serum samples from both healthy individuals and patients with cervical

carcinoma. Integrating the electrochemical biosensor into a purposely designed

electrochemical microfluidics device proved to be a very promising strategy to

improve the sensor further. We believe that, after standardization of the method,

this immunosensor should be readily adaptable to measure Nanog biomarker

expression in real patients, for point-of-care early detection and monitoring of

cancer disease.

107

We also carried out an investigation of the properties of two series of

peptide SAMs in order to check their potential as possible templates for building

the above biosensor/s. In a well-performing electrochemical biosensor, the control

of the ET through the monolayer is of paramount importance. The ET behavior is

dictated by the properties of the SAM in terms of electron tunneling and quality of

the SAM itself. Aib-homopeptides were chosen for their peculiar properties to

form stable and robust 310-helices even for short oligomers. For these systems, the

ET rate is favored by the stiffness of the secondary structure, which in turn is

associated with the presence of an extended network of intramolecular H-bonds.

Two series of thiolated Aib-homopeptides were studied: a plus series with the

positive pole of the dipole moment on the sulfur side and a minus series whose

dipole moment is reversed. In each series the number of residues was varied. For

the plus series, the number of residues was varies from 1 to 6, allowing to study

the effect of having from 0 to 5 intramolecular H-bonds (according to the 310-

helix). For the minus series we focused on peptides with 1, 3, and 5 H-bonds.

The structure and the stability of the secondary structure of the adsorbed

peptides while in the SAMs were studied by IRRAS spectroscopy. We observed

differences as a function of both the peptide length and orientation. The data

showed that stable 310-helices form for most peptides, while for the shortest

peptides competition between intra- and intermolecular H-bonds makes things

more complex. In general, we found that when peptides are organized in the

SAM, a wide network of lateral interaction is established, which contributes to

increase the stability of the SAM itself and strengthens the secondary structure of

the SAM constituents. This is particularly true for the plus series that proved to be

slightly more stable, as helices, than the corresponding series of free peptides in

solution.

Complementary information on the organization of the SAM was gathered

by calculation of the tilt angle. We found that the plus series gives SAMs with

more vertically oriented molecules and thus thicker SAMs than for the minus

series. It thus appears that the reversed dipole moment of the minus series is

somehow slightly destabilizing the structure of the SAM relative to the plus

series. Because of this destabilization, the peptides of the minus series prefer to

108

adopt a more pronounced tilt angle with respect to the surface normal. With 1- the

310-helical structure is totally modified (or even lost) in favor of later interactions.

From our study, we conclude that long peptides of the plus series are liable

to provide the most stable and ordered monolayers, whereas shorter peptides form

SAMs that are less stable and more prone to defects. This evidence, together with

the excellent ET properties of the long peptides of the plus series,89 thus suggests

that sufficiently long peptides of the plus series may provide a particularly

convenient support for the construction of efficient electrochemical biosensors.

This Thesis was made possible by the collaboration (and friendship)

between my supervisor, Prof. Flavio Maran, and co-supervisor, Prof. Jim Rusling.

The research proceeded smoothly and although some aspects are still waiting to

be addressed, the general outcome of the two related topics was as all of us hoped.

New challenges are, however, around the corner. To improve the efficiency of the

biosensor, we are about to optimize the immunosensor by changing some

elements of the transducer, particularly by using Aib-peptide SAMs which should

lead to very fast electrode kinetics. We are also investigating the possibility of

increasing the active surface of the electrode substrate, for example by using

nanostructured gold electrodes113 as an alternative to a bed of AuNPs. Some

attempts in this direction have been carried out, but this an aspect that for time

limitations of this Thesis could not be covered.

109

5. References

[1] McNaught, A. D.; Wilkinson, A.; Compendium of Chemical Terminology,

2nd ed.; Blackwell Scientific Pubblications, Oxford, 1997.

[2] Wang, J.; Glucose Biosensors: 40 years of advances challenges

Electroanalysis, 2001, 13, 983-988.

[3] Turner, A. P. F.; Karube, I.; Wilson, G. S.; Biosensors: Fundamentals and

Applications, Oxford University Press, 1987.

[4] Bashir, D. R.; Ferrari, M.; Wereley, S.; Biomolecular sensing, processing

and analysis, Springer, 1987.

[5] Chaki, N. K.; Vijayamohanan, K.; Biosensors and Bioelectronics, 2002,

17, 1-12.

[6] Koncki, R.; Anal. Chim. Acta, 2007, 599(1), 7-15.

[7] Ligler, F. S.; Anal. Chem., 2009, 81(2), 519-26.

[8] Ramanathan, K.; Danielsson, B.; Biosensor Bioelectron., 2001, 16(6), 417-

23.

[9] Mani, W.; Chikkaveeraiah, B. V.; Patel, V.; Gutkind, J. S.; Rusling, J. F.;

ACSNano, 2009, 3(3), 585-594.

[10] Pui-yee, C. C.; Cheung, C. Y.; Renneberg, R.; Seydack, M.; Biochem.

Engin./Biotechnol., 2008, 109, 123-154.

[11] Pober, J. S.; Lichtman, A. H.; Abbas, A. K.; Immunologia cellulare e

molecolare, 2002, XI-596 p., 4th Ed., Piccin-Nuova Libraria.

[12] Ambrosi, A.; Castañeda, M. T.; Killard, A. J.; Smyth, M. R.; Alegret, S.;

Merkoçi, A.; Anal. Chem., 2007, 79(14), 5232-40.

[13] De la Encosura-Muñiz, A. ; Maltez-da Costa, M.; Merkoçi, A.; Biosensors

and Bioelectronics, 2009, 24, 2475-2482.

[14] Chikkaveeraiah, B. V.; Liu, H; Mani V.; Papadimitrakopoulos, F.;

Rusling, J. F. Electrochemistry Communications, 2009, 11, 819-822.

[15] Ye, F.; Zhou, C.; Cheng, Q.; Shen, J.; Chen, H.; BioMed Central Cancer,

2008, 8:108.

110

[16] (a) Willner, I.; Katz, E. Angew. Chem., Int. Ed. 2000, 39, 1181-1218. (b)

Gooding, J. J.; Mearns, F.; Yang, W; Liu, J. Electroanalysis 2003, 15, 81-

96. (c) Rosi, N. L.; Mirkin, C. A. Chem. Rev. 2005, 105, 1547-1562. (d)

Rusling, J. F.; Wang, B.; Yun, S.-E. Electrochemistry of Redox Enzymes,

In Bioelectrochemistry, Bartlett, P. N. Ed., John Wiley, N. Y. 2008,

Chapter 2, pp. 39-86. (e) Rusling, J. F.; Hvastkovs, E. G.; Hull, D. O.;

Schenkman, J. B. Chem Commun. 2008, 141-154. (f) Polsky, R.; Harper, J.

C.; Wheeler, D. R.; Brozik, S. M. Electroanalysis 2008, 20, 671-679. (g)

Tkac, J.; Davis, J. J. Label-free Field Effect Protein Sensing, In

Engineering the Bioelectronic Interface, Davis, J. J. Ed., Royal Soc. Chem.

UK, 2009, pp. 193-224. (h) Polsky, R.; Harper, J. C.; Brozik, S. M.

Nanoparticle and Biomolecular-Nanoparticle Hybrid Supramolecular

Complexes for Electrochemical Signaling, In The Supramolecular

Chemistry of Organic-Inorganic Hybrid Materials, Rurack, K., Martinez-

Manez, R. Eds, John Wiley, N. Y. 2010, Chapter 8, pp 273-296.

[17] (a) Ulman, A. Chem. Rev. 1996, 96, 1533-1554. (b) Adams, D. M.; Brus,

L.; Chidsey, C. E. D.; Creager, S.; Creutz, C.; Kagan, C. R.; Kamat, P. V.;

Lieberman, M.; Lindsay, S.; Marcus, R. A.; Metzger, R. M.; Michel-

Beyerle, M. E.; Miller, J. R.; Newton, M. D.; Rolison, D. R.; Sankey, O.;

Schanze, K. S.; Yardley, J.; Zhu, X. J. Phys.Chem. B 2003, 107, 6668-

6697. (c) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.;

Whitesides, G. M. Chem. Rev. 2005, 105, 1103-1169. (d) Chen, D.; Li, J.

Surface Sci. Reports 2006, 61, 445-463. (e) Newton, M. D.; Smalley, J. F.

Phys. Chem. Chem. Phys. 2007, 9, 555-572.

[18] (a) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265-322.

(b) Barbara, P. F.; Meyer, T. J.; Ratner, M. A. J. Phys. Chem. 1996, 100,

13148-13168.

[19] (a) Newton, M. D. Chem. Rev. 1991, 91, 767-792. (b) Nitzan, A. Ann. Rev.

Phys. Chem. 2001, 52, 681-750. (c) Paddon-Row, M. N. Aus. J. Chem.

2003, 56, 729-748.

[20] (a) Skourtis, S. S.; Beratan, D. N. Adv. Chem. Phys. 1999, 106, 377-452.

(b) Regan, J. J.; Onuchic, J. N. Adv. Chem. Phys. 1999, 107, 497-553. (c)

Winkler, J. R. Curr. Opin. Chem. Biol. 2000, 4, 192-198. (d) Gray, H. B.;

111

Winkler, J. R. In Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-

VCH: Weinheim, 2001; Vol. 1, pp 3-23.

[21] (a) Page, C. C.; Moser, C. C.; Chen, X.; Dutton, P. L. Nature 1999, 402,

47-52. (b) Kelley, S. O.; Barton, J. K. Science 1999, 283, 375-381. (c)

Jortner, J.; Bixon, M.; Langenbacher, T.; Michel-Beyerle, M. E. Proc. Nat.

Acad. Sci. 1998, 95, 12759-12765. (d) Schuster, G. B. Acc. Chem. Res.

2000, 33, 253-260. (e) Giese, B. Ann. Rev. Biochem. 2002, 71, 51-70. (f)

Berlin, Y. A.; Burin, A. L.; Ratner, M. A. Chem. Phys. 2002, 275, 61-74.

(g) Lewis, D.; Liu, J.; Weigel, W.; Rettig, W.; Kurnikov, I. V.; Beratan, D.

N. Proc. Nat. Acad. Sci. 2002, 99, 12536-12541.

[22] a) Anderson, P. W. Phys. Rev. 1950, 79, 350-356. (b) McConnell, H. M. J.

Chem. Phys. 1961, 35, 508-515.

[23] (a) Grozema, F. C.; Berlin, Y. A.; Siebbeles, L. D. A. J. Am. Chem. Soc.

2000, 122, 10903-10909. (b) Bixon, M.; Jortner, J. J. Am. Chem. Soc.

2001, 123, 12556-12567. (c) Li, X.-Q.; Zhang, H.; Yan, Y. J. J. Phys.

Chem. A 2001, 105, 9563-9567. (d) Renger, T.; Marcus, R. A. J. Phys.

Chem. A 2003, 107, 8404-8419.

[24] (a) Long, Y.-T.; Abu-Irhayem, E.; Kraatz, H.-B. Chem. Eur. J. 2005, 11,

5186-5194. (b) Schlag, E. W. , Sheu, S.-Y. ; Yang, D.-Y. ; Selzle, H. L.;

Lin, S. H. Angew. Chem., Int. Ed. 2007, 46, 3196-3210.

[25] (a) Yanagisawa, K.; Morita, T.; Kimura, S. J. AM. Chem. Soc. 2004, 126,

12780-12781. (c) Giese, B.; Graber, M.; Cordes, M. Curr. Opin. Chem.

Biol. 2008, 12, 755-759.

[26] Antonello, S.; Maran, F. Chem. Soc. Rev. 2005, 34, 418-428.

[27] Rampi, M. A.; Whitesides, G. M. Chem. Phys. 2002, 281, 373-391.

[28] Smalley, J. F.; Finklea, H. O.; Chidsey, C. E. D.; Linford, M. R.; Creager,

S. E.; Ferraris, J. P.; Chalfant, K.; Zawodzinsk, T.; Feldberg, S. W.;

Newton, M. D. J. Am. Chem. Soc. 2003, 125, 2004-2013.

[29] (a) Morita, T.; Kimura, S. J. Am. Chem. Soc. 2003, 125, 8732-8733. (b)

Watanabe, J.; Morita, T.; Kimura, S. J. Phys. Chem. B 2005, 109, 14416–

14425. (c) Kai, M.; Takeda, K.; Morita, T.; Kimura, S. J. Pept. Sci. 2008,

14, 192–202. (d) Takeda, K.; Morita, T.; Kimura, S. J. Phys. Chem. B

112

2008, 112, 12840–12850. (e) Arikuma, Y.; Takeda, K.; Morita, T.; Ohmae,

M.; Kimura, S. J. Phys. Chem. B 2009, 113, 6256–6266.

[30] (a) Galka, M. M.; Kraatz, H.-B. ChemPhysChem 2002, 3, 356-359. (b)

Bediako-Amoa, I.; Sutherland, T. C.; Li, C.-Z.; Silerova, R.; Kraatz, H.-B.

J. Phys. Chem. B. 2004, 108, 704-714. (c) Kraatz, H.-B.; Bediako-Amoa,

I.; Gyepi-Garbrah, H.; Sutherland, T. C. J. Phys. Chem. B. 2004, 108,

20164-20172. (d) Dey, S. K.; Long, Y. T.; Chowdhury, S.; Sutherland, T.

C.; Mandal, H. S.; Kraatz, H. B. Langmuir 2007, 23, 6475–6477. (e)

Orlowski, G. A.; Chowdhury, S.; Kraatz, H. B. Langmuir 2007, 23,

12765–12770.

[31] (a) Bilewicz, R.; Sek, S.; Zawiska, I. Russ. J. Electrochem. 2002, 38, 29-

38. (b) Sek, S.; Moszynski, R.; Sepiol, A.; Misicka, A.; Bilewicz, R. J.

Electroanal. Chem. 2003, 550-551, 359-364. (c) Sek, S.; Sepiol, A.; Tolak,

A; Misicka, A.; Bilewicz, R. J. Phys. Chem. B 2004, 108, 8102-8105. (d)

Sek, S.; Tolak, A.; Misicka, A.; Palys, B.; Bilewicz, R. J. Phys. Chem. B

2005, 109, 18433–18438. (e) Sek, S.; Swiatek, K.; Misicka, A. J. Phys.

Chem. B 2005, 109, 23121–23124. (f) Sek, S.; Misicka, A.; Swiatek, K.;

Maicka, E. J. Phys. Chem. B 2006, 110, 19671–19677. forse solo SPM

[32] For recent ET studies on redox proteins and relevant references, see: (a)

Davis, K. L.; Drews, B. J.; Yue, H.; Waldeck, D. H.; Knorr, K.; Clark, R.

A. J. Phys. Chem. C 2008, 112, 6571–6576. (b) Millo, D.; Ranieri, A.;

Gross, P.; Ly, H. K.; Borsari, M.; Hildebrandt, P.; Wuite, G. J. L.; Gooijer,

C.; van der Zwan, G. J. Phys. Chem. C 2009, 113, 2861-2866.

[33] Brooksby, P. A.; Anderson, K. H.; Downard, A. J.; Abell, A. D. Langmuir,

2010, 26, 1334-1339.

[34] Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem.

Soc. 1987, 109, 3559-3568.

[35] (a) Becka, A. M.; Miller, C. J. J. Am. Chem. Soc. 1992, 96, 2657-2668. (b)

Becka, A. M.; Miller, C. J. J. Phys. Chem. 1992, 96, 2657-2668. (c) Becka,

A. M.; Miller, C. J. J. Phys. Chem. 1993, 97, 6233-6239. (d)Terrettaz, S.;

Becka, A. M.; Traub, M. J.; Fettinger, J. C.; Miller, C. J. J. Phys. Chem.

1995, 99, 11216-11224. (e) Cheng, J.; Sàghi-Szabò, G.; Tossel, J. A.;

Miller, C. J. J. Am. Chem. Soc.1996, 118, 680-684.

113

[36] Krysinski, P.; Brostowska-Smolska, M. J. Electroanal Chem. 1997, 424,

61-67.

[37] Doneux, T.; Steichen, M.; Bouchta, T.; Buess-Herman, C. J. Electroanal.

Chem. 2007, 599, 241-248.

[38] Protsailo, L. V.; Fawcett W. R. Electrochim. Acta 2000, 45, 3497–3505.

(Studio tramite impedenza)

[39] Smalley, J. F.; Newton, M. D.; Feldberg, S. W. J. Electroanal. Chem.

2006, 589, 1–6.

[40] (a) Degefa, T. H.; Schön, P.; Bongard, D.; Walder, L. J. Electroanal.

Chem. 2004, 574, 49-62. (b) Hwang, S.; Lee, B. S.; Chi, Y. S.; Kwak, J.;

Choi, I. S.; Lee, S. J. Electrochim. Acta 2008, 53, 2630-2636.

[41] Badia, A.; Lennox, B.; Reven, L. Acc. Chem. Res. 2000, 33, 475-481.

[42] Bizzotto, D.; Yu, H.; Cheng, A.K.H.; Murphy, J. N. J. Am. Chem. Soc.,

2009, 131, 4042-4050.

[43] (a) Zhao, X.; Wilbur, J. L.; Whitesides, G. M. Langmuir 1996, 12, 3257-

3264. (b) Che, G.; Cabrera C. R. J. Electroanal. Chem. 1996, 417, 155-

161. (c) Che, G.; Li, Z.; Zhang, H.; Cabrera C. R. J. Electroanal. Chem.

1998, 453, 9–17. (d) Protsailo, L. V.; Fawcett, W. R. Langmuir 2002, 18,

8933-8941. (Analisi dei difetti con Impedenza) (e) Benìtez, G.; Vericat,

C.; Tanco, S.; Lenicov, R. F.; Castez, M. F.; Vela, M. E.; Salvarezza R. C.;

Langmuir 2004, 20, 5030-5037. (f) Kiani, A.; Alpuche-Aviles, M. A.;

Eggers, P. K.; Jones, M.; Gooding, J. J.; Paddon-Row, M. N.; Bard, A. J.

Langmuir 2008, 24, 2841-2849. (g) Preiner, M. J.; Melosh, N. A.

Langmuir 2009, 25, 2585-2587.

[44] (a) Liu, B.; Mirkin, M. V.; Bard, A. J.; Creager, S. E. J. Am. Chem. Soc.

2004, 126, 1485-1492.; (b) Finklea, O. H. Langmuir 1993, 9, 3660-3667;

(c) Chailapaku, O.; Crooks, R. M. Langmuir 1995, 11, 1329-1340; (d)

Orlowski, G. A..; Kraatz, H. Electrochimica Acta 2006, 51, 2934–2937.

[45] Strattmann, M. Adv. Mater. 1990, 2, 191.

[46] Ulman, A. Thin Solid Films 1996, 273, 48-53.

[47] Nuzzo, R. C.; Dubois, L. H.; Allara, D. L. J. Am. Chem. Soc. 1990, 112,

558-569.

[48] DePalma, V.; Tillman, N. Langmuir 1989, 5, 868-872.

114

[49] Ferguson, G. S.; Chaudhury, M. K.; Sigal, G. B.; Whitesides, G. M.

Science 1991, 253, 776-778.

[50] Chen, J.; Reed, M. A.; Tour, J. M. Science 1990, 286, 1550-1552.

[51] Katz, H. E.; Wilson, W. L.; Scheller, G. J. Am. Chem. Soc. 1994, 116,

6636-6640.

[52] Malem, F.; Mandler, D. Anal. Chem. 1993, 65, 37-41.

[53] Gooding, J. J.; Praig, V. G.; Hall, E. A. H. Anal. Chem. 1998, 70, 2396-

2402.

[54] Giz, M. J.; Duong, B.; Tao, N. J. J. Electroanal. Chem. 1999, 465, 72-79

[55] Chen, D.; Li, J. Surface Sci. Reports 2006, 61, 445-463.

[56] Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G.M.

Chem. Rev. 2005, 105, 1103-1169.

[57] Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. J. Am. Chem. Soc. 1987, 109,

733-740.

[58] Valiokas, R.; Svedhem, S.; Svensson, S. C. T.; Liedberg, B.

Langmuir 1999, 15, 3390-3394.

[59] Schlenoff, J. B.; Li, M.; Ly, H. J. Am. Chem. Soc. 1995, 117, 12528.

[60] Antonello, S.; Benassi, R.; Gavioli, G.; Taddei, F.; Maran, F. J. Am. Chem.

Soc. 2002, 124, 7529-7538.

[61] Poirier, G.E. Langmuir 1997, 13, 2019-2026.

[62] Paradis, E.; Rowtree, P. J. Electroanal. Chem. 2003, 550, 175-185.

[63] Zhang, L.; Goddard, W. A.; Jiang, S. J. Chem. Phys. 2002, 117, 7341-

7349.

[64] Jiang, S. Mol. Phys. 2002, 100, 2261-2275.

[65] Fischer, D.; Curioni, A.; Andreoni, W. Langmuir 2003, 19, 3567-3571.

[66] Heister, K.; Zharnikov, M.; Grunze, M.; Johansson, L. S. O. J. Phys.

Chem. B 2001, 105, 4058-4061.

[67] Yang, Y. W.; Fan, L. J. Langmuir 2002, 18, 1157-1164.

[68] Dubois, L. H.; Zegarski, B. R.; Nuzzo, R.G. J. Chem. Phys. 1993, 98, 178-

188.

[69] Camillone, N.; Chidsey, C. E. S.; Eisemberger, P.; Fenter, P.; Li, J.; Liang,

K. S.; Liu, G. Y.; Scoles, G. J. Chem. Phys. 1993, 99, 744-747.

115

[70] Fenter, P; Eisemberger, P.; Liang, K. S. Phys. Rev. Lett. 1993, 70, 2447-

2450.

[71] Seydack, M., Biosens Bioelectron (2005) 20, 2454.

[72] Roy, S., and Gao, Z., Nano Today (2009) 4, 318.

[73] Pumera, M., et al., Sensors and Actuators B (2007) 123, 1195.

[74] Rosi, N. L., and Mirkin, C. A., Chem Rev (2005) 105, 1547.

[75] De la Enconsura-Muñiz A, Parolo C., Merkoçi A, Materialstoday (2010),

13.

[76] De la Enconsura-Muñiz, A.; Trends Anal Chem (2008) 27 (7), 568.

[77] Karle, I. L.; Balaram, P. Biochemistry 1990, 29, 6747-6756.

[78] (a) Toniolo, C.; Crisma, M.; Formaggio, F.; Peggion, C. Biopolymers

2001, 60, 396-419. (b) Crisma, M.; Moretto, A.; Formaggio, F.; Toniolo,

C. Biopolymers 2006, 84, 3-12.

[79] Goodman, M.; Toniolo, C.; Pallai, P. In Forum Peptides; Castro, B.,

Martinez, J., Eds.; Nancy Dohr: France, 1985; pp 146-174.

[80] (33) Toniolo, C.; Bonora, G. M.; Barone, V.; Bavoso, A.; Benedetti, E.; Di

Blasio, B.; Grimaldi, P.; Lelj, F.; Pavone, V.; Pedone, C., Macromolecules

1985, 18, 895-902.

[81] (a) Improta, R.; Barone, V.; Kudin, K. N.; Scuseria, G. E. J. Am. Chem.

Soc. 2001, 123, 3311-3322. (b) Shin, Y.-G; Newton, M. D.; Isied, S. S. J.

Am. Chem. Soc. 2003, 125, 3722-3732. (c) Wieczorek, R.; Dannenberg, J.

J. J. Am. Chem. Soc. 2004, 126, 14198-14205.

[82] (a) Kennedy, D. F.; Crisma, M.; Toniolo, C.; Chapman, D. Biochemistry

1991, 30; (b) Hanson, P.; Millhauser, F.; Formaggio, F.; Crisma, M.;

Toniolo, C. J. Am. Chem. Soc. 1996, 118.

[83] (a) Fabris, L.; Antonello, S.; Armelao, L.; Donkers, R. L.; Polo, F.;

Toniolo, C.; Maran, F., J. Am. Chem. Soc. 2006, 128, 326-336. (b) Holm,

A. H.; Ceccato, M.; Donkers, R. L.; Fabris, L.; Pace, G.; Maran, F.,

Langmuir 2006, 22, 10584-10589.

[84] Perera, N. V.; Isley, W.; Maran, F.; Gascón, J. A. J. Phys. Chem. C 2010,

114, 1643-16050.

[85] Wen, X. G.; Linton, R. W.; Formaggio, F.; Toniolo, C.; Samulski, E. T. J.

Phys. Chem. A 2004, 108, 9673-9681.

116

[86] a) Antonello, S.; Formaggio, F.; Moretto, A.; Toniolo, C.; Maran, F. J. Am.

Chem. Soc. 2003, 125, 2874-2875. (b) Polo, F.; Antonello, S.; Formaggio,

F.; Toniolo, C.; Maran, F. J. Am. Chem. Soc. 2005, 127, 492-493.

[87] (a) Improta, R.; Antonello, S.; Formaggio, F.; Maran, F.; Rega, N.;

Barone, V. J. Phys. Chem. B 2005, 109, 1023-1033. (b) Improta, R.;

Barone, V.; Newton, M. D.; Chemphyschem, 2006, 7, 1211.

[88] Becucci, L.; Guryanov, I.; Maran, F.; Guidelli, R. J. Am. Chem. Soc. 2010,

132, 6194-6204.

[89] Hesari, M.; Antonello, S.; Armelao, L.; Gobbo, P.; Guidelli, R.; Guryanov,

I.; Rampi, M. A.; Soldà, A.; Maran, F. Manuscript in preparation.

[90] Lvov, Y.M Thin-Film Nanofabrication by alterbate Absorption of

Polyions, Nanoparticles, and Proteins. In Handbook of Surface and

Interface of Material: Nanostructured Materials, Micelles, and Colloids;

Nalwa, H.S, Ed.; Academic Press: San Diego, 2001; Vol. 3, 170-189.

[91] Angerstein-Kozlowska, H.; Conway, B. E.; Hamelin, A.; Stoicoviciu, L. J.

Electroanal. Chem. 1987, 228, 429-453. Tkac, J.; Davis, J. J. J.

Electroanal. Chem. 2008, 621, 117-120.

[92] Haiss, W.; Thanh, N. T. K.; Aveyard, J.; Fernig, D. G.; Anal. Chem. 2007,

79, 4215-4221.

[93] Galus Z.; Fundamentals of electrochemical analysis, 2nd Edition, Polish

Scientific Publishers Pwn, 1976.

[94] Juntao Lu and Peifang Liu, J. Electroanal. Chem., 1988, 244, 339.

[95] Engstrom, R. C.; Weber, M.; Wunder, D. J.; Burgess, R.; Winquist, S.;

Anal. Chem. 1986, 58, 844.

[96] Tanaka, K.; Tamamushi, R.; Bioelectrochem. Bioenerg. 1987, 17, 519.

[97] Tanaka, K.; Kashiwagi, N.; Bioelectrochem. Bioenerg. 1988, 20, 1.

[98] Zamlynni, V. PhD Thesis Work, “Electrochemical and Spectroscopic

Studies of Pyridine Surfactants at the Gold-Electrolyte Interface”, 2002,

University of Guelph, ON Canada.

[99] Sheppard, N.; Erkelens, J. Appl. Spec. 1984, 38, 471-485.

[100] Toniolo, C.; Bonora, G. M.; Barone, V.; Bavoso, A.; Benedetti, E.; Di

Blasio, B.; Grimaldi, P.; Lelj, F.; Pavone, V.; Pedone, C.;

Macromolecules 1985, 18, 895-902

117

[101] Buffeteau, T.; Desbat, B., Appl. Spec. 1989, 6, 1027-1032.

[102] Strong, L.; Whitesides, G. M., Langmuir 1988, 4, 546-558.

[103] Swalen, J. D.; Allara, D. L.; Andrade, J. D.; Chandross, E. A.; Garoff, S.;

Israelachvili, J.; McCrthy, T. J.; Murray, R.; Pease, R. F.; Rabolt, J. F.,

Wynne, K. J.; Yu, H.; Langmuir 1987, 3, 932.

[104] (a) Wen, X.; Linton, R.W.; Formaggio, F.; Toniolo, C.; Samulski, E.T.; J.

Phys. Chem. A 2004; (b) Enriquez, E. P.; Samulski, E.T. Mat. Res. Soc.

Symp. Proc. 1992, 255, 423.

[105] Greenler, R. G., J. Chem. Phys. 1966, 44, 310-315.

[106] Michl, J.; Thulstrup, E. W.; Spectroscopy with Polarized Light: Solute

alignment by Photoselection, in Liquid Crystal, Polymers, and

Menmbranes; VCH Publichers: Deerfield Beach, FL, 1986.

[107] Tsuboi, M.; J. Polymer Sci. 1962, 59, 139-153.

[108] Arikuma, Y.; Takeda, K.; Morita, T.; Ohmae, M.; Kimura, S. J. Phys.

Chem. B 2009, 113, 6256–6266.

[109] Kraatz, H.-B.; Bediako-Amoa, I.; Gyepi-Garbrah, H.; Sutherland, T. C. J.

Phys. Chem. B. 2004, 108, 20164-20172.

[110] Mandal, H. S.; Kraatz, H. B. Chem. Phys. 2006, 326, 246–251.

[111] (a) Kitagawa, K.; Morita, T.; Kimura, S. Thin Solid Films 2006, 509, 18–

26. (b) Watanabe, J.; Morita, T.; Kimura, S. J. Phys. Chem. B 2005, 109,

14416-14425. (c) Kitagawa, K.; Morita, T.; Kimura, S. J. Phys. Chem. B

2004, 108, 15090-15095.

[112] Ward, M. A., Catto, J.W.F., Hamdy, F.C., Ann. Clin. Biochem, 2001, 38,

633-651.

[113] Cattarin, S.; Kramer, D.; Lui, A.; Musiani, M.; J. Phys. Chem. C 2007,

111, 12643-12649.

electrochemical immunosensors and peptide self …tesi.cab.unipd.it/26611/1/solda_tesi.pdf ·...

Documents